# Estimate with missing values on dependent variable?

**URL:** https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96
**Category:** Estimation
**Created:** [December 3, 2019, 12:04am UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96 "2019-12-03T00:04:39Z")
**Posts on this page:** 13
**Page:** 1

<div class="post-metadata">

### Author: ![alewolf](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/alewolf/32/24_2.png) [@alewolf](https://discourse.pumas.ai/u/alewolf)
#### Post date: [December 3, 2019, 12:04am UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/1 "2019-12-03T00:04:39Z")

</div>

I’ve been working on a fitting where there are some measurements before dosing, but not at the exact dosing time -

```auto
time tad rate amt Drug
Float64 Float64	Float64⍰	Float64⍰	Float64⍰
0.0 0.0 missing missing 12
0.1 0.0 1000 100.0 missing
0.5 0.4 missing	missing 1000
1.0 0.9 missing	missing 980

```

evid is 1 where amt is not missing, 0 otherwise.

I define my julia dataset as:

```auto

dfP=read_pumas(df9 ,id=:Subject, evid=:evid, time=:Time, amt=:Dosemg, rate=:rate, dvs=[:Drug])

```

In the model

```auto
@derived begin
        a1 = A1
        Conc_Mean = C1
        Drug ~ @. Normal(Conc_Mean, sqrt(Conc_Mean^2*σ_prop))

```

However when I run the estimation my model fails:

```auto
param=init_param(Model)
# res3 = fit(Model,dfP,param,Pumas.FOCEI())

res2= fit(Model,dfP,param,Pumas.NaivePooled())

```

I wonder if this is because of my missing value in :Drug, which is my dependent variable or if something else is going on?  
If the former - is there a way for the estimator to inpute the value?

┌ Warning: First function call produced NaNs. Exiting.  
└ @ OrdinaryDiffEq C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\OrdinaryDiffEq\UsWa2\src\initdt.jl:135  
┌ Warning: Automatic dt set the starting dt as NaN, causing instability.  
└ @ OrdinaryDiffEq C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\OrdinaryDiffEq\UsWa2\src\solve.jl:403

┌ Warning: First function call produced NaNs. Exiting.  
└ @ OrdinaryDiffEq C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\OrdinaryDiffEq\UsWa2\src\initdt.jl:135

**type NamedTuple has no field dv**

Stacktrace:  
[1] getproperty(::Any, ::Symbol) at .\Base.jl:20  
[2] (::getfield(Pumas, Symbol("##462#464")))(::Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\6uorK\src\estimation\show.jl:15  
[3] \_mapreduce(::getfield(Pumas, Symbol("##462#464")), ::typeof(Base.add\_sum), ::IndexLinear, ::Array{Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1}) at .\reduce.jl:317  
[4] \_mapreduce\_dim(::Function, ::Function, ::NamedTuple{,Tuple{}}, ::Array{Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1}, ::Colon) at .\reducedim.jl:312  
[5] #mapreduce#555 at .\reducedim.jl:307 [inlined]  
[6] mapreduce at .\reducedim.jl:307 [inlined]  
[7] \_sum(::Function, ::Array{Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1}, ::Colon) at .\reducedim.jl:657  
[8] #sum#559 at .\reducedim.jl:653 [inlined]  
[9] sum at .\reducedim.jl:653 [inlined]  
[10] \_print\_fit\_header(::IOContext{Base.GenericIOBuffer{Array{UInt8,1}}}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1107#1114")),getfield(Main, Symbol("##1108#1115")),getfield(Main, Symbol("##1109#1116")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1110#1117")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1112#1119")),getfield(Main, Symbol("##1113#1120"))},Array{Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\6uorK\src\estimation\show.jl:14  
[11] show(::IOContext{Base.GenericIOBuffer{Array{UInt8,1}}}, ::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1107#1114")),getfield(Main, Symbol("##1108#1115")),getfield(Main, Symbol("##1109#1116")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1110#1117")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1112#1119")),getfield(Main, Symbol("##1113#1120"))},Array{Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\6uorK\src\estimation\show.jl:23  
[12] limitstringmime(::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1107#1114")),getfield(Main, Symbol("##1108#1115")),getfield(Main, Symbol("##1109#1116")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1110#1117")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1112#1119")),getfield(Main, Symbol("##1113#1120"))},Array{Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\inline.jl:37  
[13] display\_mimestring(::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1107#1114")),getfield(Main, Symbol("##1108#1115")),getfield(Main, Symbol("##1109#1116")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1110#1117")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1112#1119")),getfield(Main, Symbol("##1113#1120"))},Array{Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\display.jl:67  
[14] display\_dict(::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1107#1114")),getfield(Main, Symbol("##1108#1115")),getfield(Main, Symbol("##1109#1116")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1110#1117")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1112#1119")),getfield(Main, Symbol("##1113#1120"))},Array{Subject{NamedTuple{(:Drug,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##200#201")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\display.jl:96  
[15] #invokelatest#1 at .\essentials.jl:790 [inlined]  
[16] invokelatest at .\essentials.jl:789 [inlined]  
[17] execute\_request(::ZMQ.Socket, ::IJulia.Msg) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\execute\_request.jl:112  
[18] #invokelatest#1 at .\essentials.jl:790 [inlined]  
[19] invokelatest at .\essentials.jl:789 [inlined]  
[20] eventloop(::ZMQ.Socket) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\eventloop.jl:8  
[21] (::getfield(IJulia, Symbol("##15#18")))() at .\task.jl:268

---

<div class="post-metadata">

### Author: ![vijay](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/vijay/32/96_2.png) [@vijay](https://discourse.pumas.ai/u/vijay)
#### Post date: [December 3, 2019, 12:56am UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/2 "2019-12-03T00:56:40Z")

</div>

1. Can you try by renaming `Drug` to `dv` in your dataset
2. What is your measurement before dosing? Is this an endogenous substance? If so, is that value at steady state or did you define an initial condition for that compartment? If your answer to any of these questions is “no”, you likely will not be able to estimate from that data

---

<div class="post-metadata">

### Author: ![alewolf](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/alewolf/32/24_2.png) [@alewolf](https://discourse.pumas.ai/u/alewolf)
#### Post date: [December 3, 2019, 1:18am UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/3 "2019-12-03T01:18:46Z")

</div>

Hi Vijay,

1. did the first thing and I got other error.

I’ll check tomorrow at work and forward to you in the morning.

1. yes, but I didn’t define a steady state value or initial condition.

Should I set steady state as the value pre admin?

Thanks-

A

---

<div class="post-metadata">

### Author: ![vijay](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/vijay/32/96_2.png) [@vijay](https://discourse.pumas.ai/u/vijay)
#### Post date: [December 3, 2019, 1:23am UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/4 "2019-12-03T01:23:33Z")

</div>

If the compartment for which you have a differential equation has an amount to begin with (before dosing), then you should either estimate that amount or set it’s initial value (e.g baseline) using the `@init` block

---

<div class="post-metadata">

### Author: ![andreasnoack](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/andreasnoack/32/22_2.png) [@andreasnoack](https://discourse.pumas.ai/u/andreasnoack)
#### Post date: [December 3, 2019, 7:36am UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/5 "2019-12-03T07:36:29Z")

</div>

> [@alewolf](#):
>
> type NamedTuple has no field dv

This error suggests that you are not using the latest version of Pumas. Could you please try `] up Pumas` from the command prompt? Once you have upgraded, it should be fine to call the dependent variable `Drug`.

Regarding the

```auto
Warning: First function call produced NaNs. Exiting.

```

errors there can be various reasons.

> I wonder if this is because of my missing value in :Drug, which is my dependent variable

As such, a missing value shouldn’t be a problem. Pumas should just ignore it. One simple thing you could try would be to set `alg=Tsit5()` in `fit`. In some rare cases, we have seen issues with the default solver.

---

<div class="post-metadata">

### Author: ![alewolf](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/alewolf/32/24_2.png) [@alewolf](https://discourse.pumas.ai/u/alewolf)
#### Post date: [December 3, 2019, 5:24pm UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/6 "2019-12-03T17:24:49Z")

</div>

So, I updated Pumas and tried Vijay solution (adding an initial condition to A1 from which dv is derived, and naming :Drug :dv in the Dataframe)  
I got this error

┌ Warning: dt \<= dtmin. Aborting. There is either an error in your model specification or the true solution is unstable.  
└ @ DiffEqBase C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\DiffEqBase\IDJcQ\src\integrator\_interface.jl:170

**BoundsError: attempt to access 9-element Array{String,1} at index [0]**

Stacktrace:  
[1] getindex at .\array.jl:728 [inlined]  
[2] getindex at .\multidimensional.jl:486 [inlined]  
[3] (::getfield(Pumas, Symbol("##493#494")))(::Char) at .\none:0  
[4] iterate at .\generator.jl:47 [inlined]  
[5] collect\_to!(::Array{String,1}, ::Base.Generator{String,getfield(Pumas, Symbol("##493#494"))}, ::Int64, ::Int64) at .\array.jl:651  
[6] collect\_to\_with\_first! at .\array.jl:630 [inlined]  
[7] collect(::Base.Generator{String,getfield(Pumas, Symbol("##493#494"))}) at .\array.jl:611  
[8] \_to\_subscript at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:2 [inlined]  
[9] \_push\_varinfo!(::Array{Any,1}, ::Array{Any,1}, ::Nothing, ::Nothing, ::Symbol, ::Array{Float64,1}, ::Nothing, ::Nothing) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:118  
[10] show(::IOContext{Base.GenericIOBuffer{Array{UInt8,1}}}, ::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##113#120")),getfield(Main, Symbol("##114#121")),getfield(Main, Symbol("##115#122")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##116#123")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##118#125")),getfield(Main, Symbol("##119#126"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:28  
[11] limitstringmime(::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##113#120")),getfield(Main, Symbol("##114#121")),getfield(Main, Symbol("##115#122")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##116#123")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##118#125")),getfield(Main, Symbol("##119#126"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\inline.jl:37  
[12] display\_mimestring(::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##113#120")),getfield(Main, Symbol("##114#121")),getfield(Main, Symbol("##115#122")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##116#123")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##118#125")),getfield(Main, Symbol("##119#126"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\display.jl:67  
[13] display\_dict(::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##113#120")),getfield(Main, Symbol("##114#121")),getfield(Main, Symbol("##115#122")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##116#123")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##118#125")),getfield(Main, Symbol("##119#126"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\display.jl:96  
[14] #invokelatest#1 at .\essentials.jl:790 [inlined]  
[15] invokelatest at .\essentials.jl:789 [inlined]  
[16] execute\_request(::ZMQ.Socket, ::IJulia.Msg) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\execute\_request.jl:112  
[17] #invokelatest#1 at .\essentials.jl:790 [inlined]  
[18] invokelatest at .\essentials.jl:789 [inlined]  
[19] eventloop(::ZMQ.Socket) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\eventloop.jl:8  
[20] (::getfield(IJulia, Symbol("##15#18")))() at .\task.jl:268

---

<div class="post-metadata">

### Author: ![alewolf](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/alewolf/32/24_2.png) [@alewolf](https://discourse.pumas.ai/u/alewolf)
#### Post date: [December 3, 2019, 5:27pm UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/7 "2019-12-03T17:27:08Z")

</div>

Then I tried to use alg =Tsit5()

and got the following error:

Spoilers common elemnt of error is

**BoundsError: attempt to access 9-element Array{String,1} at index [0]** -\> Though I am not sure what string array it is referring to

┌ Warning: Interrupted. Larger maxiters is needed.  
└ @ DiffEqBase C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\DiffEqBase\IDJcQ\src\integrator\_interface.jl:156

**BoundsError: attempt to access 9-element Array{String,1} at index [0]**

Stacktrace:  
[1] getindex at .\array.jl:728 [inlined]  
[2] getindex at .\multidimensional.jl:486 [inlined]  
[3] (::getfield(Pumas, Symbol("##493#494")))(::Char) at .\none:0  
[4] iterate at .\generator.jl:47 [inlined]  
[5] collect\_to!(::Array{String,1}, ::Base.Generator{String,getfield(Pumas, Symbol("##493#494"))}, ::Int64, ::Int64) at .\array.jl:651  
[6] collect\_to\_with\_first! at .\array.jl:630 [inlined]  
[7] collect(::Base.Generator{String,getfield(Pumas, Symbol("##493#494"))}) at .\array.jl:611  
[8] \_to\_subscript at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:2 [inlined]  
[9] \_push\_varinfo!(::Array{Any,1}, ::Array{Any,1}, ::Nothing, ::Nothing, ::Symbol, ::Array{Float64,1}, ::Nothing, ::Nothing) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:118  
[10] show(::IOContext{Base.GenericIOBuffer{Array{UInt8,1}}}, ::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##141#148")),getfield(Main, Symbol("##142#149")),getfield(Main, Symbol("##143#150")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##144#151")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##146#153")),getfield(Main, Symbol("##147#154"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Symbol,Tsit5,Tuple{Symbol},NamedTuple{(:alg,),Tuple{Tsit5}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:28  
[11] limitstringmime(::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##141#148")),getfield(Main, Symbol("##142#149")),getfield(Main, Symbol("##143#150")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##144#151")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##146#153")),getfield(Main, Symbol("##147#154"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Symbol,Tsit5,Tuple{Symbol},NamedTuple{(:alg,),Tuple{Tsit5}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\inline.jl:37  
[12] display\_mimestring(::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##141#148")),getfield(Main, Symbol("##142#149")),getfield(Main, Symbol("##143#150")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##144#151")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##146#153")),getfield(Main, Symbol("##147#154"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Symbol,Tsit5,Tuple{Symbol},NamedTuple{(:alg,),Tuple{Tsit5}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\display.jl:67  
[13] display\_dict(::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##141#148")),getfield(Main, Symbol("##142#149")),getfield(Main, Symbol("##143#150")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##144#151")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##146#153")),getfield(Main, Symbol("##147#154"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Symbol,Tsit5,Tuple{Symbol},NamedTuple{(:alg,),Tuple{Tsit5}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\display.jl:96  
[14] #invokelatest#1 at .\essentials.jl:790 [inlined]  
[15] invokelatest at .\essentials.jl:789 [inlined]  
[16] execute\_request(::ZMQ.Socket, ::IJulia.Msg) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\execute\_request.jl:112  
[17] #invokelatest#1 at .\essentials.jl:790 [inlined]  
[18] invokelatest at .\essentials.jl:789 [inlined]  
[19] eventloop(::ZMQ.Socket) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\eventloop.jl:8  
[20] (::getfield(IJulia, Symbol("##15#18")))() at .\task.jl:268

---

<div class="post-metadata">

### Author: ![vijay](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/vijay/32/96_2.png) [@vijay](https://discourse.pumas.ai/u/vijay)
#### Post date: [December 3, 2019, 6:45pm UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/8 "2019-12-03T18:45:15Z")

</div>

Can you paste your model here? Did you change to `dv` in the `@derived` block?

---

<div class="post-metadata">

### Author: ![alewolf](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/alewolf/32/24_2.png) [@alewolf](https://discourse.pumas.ai/u/alewolf)
#### Post date: [December 3, 2019, 9:07pm UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/9 "2019-12-03T21:07:48Z")

</div>

yes -  
I did.

This is a sensitive model. I could get in trouble by pasting it publicly. Can I share it directly with you privately if needed?

I remove the early times (before dosing) and subtracted the values as background, and it seems that solved the issue, Julia is bee running for 30 mins without crashing - so fingers crossed.  
I am still confused how this would work if I have not done that.

Take care,

A

---

<div class="post-metadata">

### Author: ![alewolf](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/alewolf/32/24_2.png) [@alewolf](https://discourse.pumas.ai/u/alewolf)
#### Post date: [December 3, 2019, 9:38pm UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/10 "2019-12-03T21:38:04Z")

</div>

Can I share with you directly

I got the same error - even with the corrected database

**BoundsError: attempt to access 9-element Array{String,1} at index [0]**

Stacktrace:  
[1] getindex at .\array.jl:728 [inlined]  
[2] getindex at .\multidimensional.jl:486 [inlined]  
[3] (::getfield(Pumas, Symbol("##493#494")))(::Char) at .\none:0  
[4] iterate at .\generator.jl:47 [inlined]  
[5] collect\_to!(::Array{String,1}, ::Base.Generator{String,getfield(Pumas, Symbol("##493#494"))}, ::Int64, ::Int64) at .\array.jl:651  
[6] collect\_to\_with\_first! at .\array.jl:630 [inlined]  
[7] collect(::Base.Generator{String,getfield(Pumas, Symbol("##493#494"))}) at .\array.jl:611  
[8] \_to\_subscript at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:2 [inlined]  
[9] \_push\_varinfo!(::Array{Any,1}, ::Array{Any,1}, ::Nothing, ::Nothing, ::Symbol, ::Array{Float64,1}, ::Nothing, ::Nothing) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:118  
[10] show(::IOContext{Base.GenericIOBuffer{Array{UInt8,1}}}, ::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1257#1264")),getfield(Main, Symbol("##1258#1265")),getfield(Main, Symbol("##1259#1266")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1260#1267")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1262#1269")),getfield(Main, Symbol("##1263#1270"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Symbol,Tsit5,Tuple{Symbol},NamedTuple{(:alg,),Tuple{Tsit5}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\Pumas\0Bex7\src\estimation\show.jl:28  
[11] limitstringmime(::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1257#1264")),getfield(Main, Symbol("##1258#1265")),getfield(Main, Symbol("##1259#1266")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1260#1267")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1262#1269")),getfield(Main, Symbol("##1263#1270"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Symbol,Tsit5,Tuple{Symbol},NamedTuple{(:alg,),Tuple{Tsit5}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\inline.jl:37  
[12] display\_mimestring(::MIME{Symbol(“text/plain”)}, ::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1257#1264")),getfield(Main, Symbol("##1258#1265")),getfield(Main, Symbol("##1259#1266")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1260#1267")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1262#1269")),getfield(Main, Symbol("##1263#1270"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Symbol,Tsit5,Tuple{Symbol},NamedTuple{(:alg,),Tuple{Tsit5}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\display.jl:67  
[13] display\_dict(::Pumas.FittedPumasModel{PumasModel{ParamSet{NamedTuple{(:θ, :σ\_prop),Tuple{VectorDomain{Array{Float64,1},Array{TransformVariables.Infinity{true},1},Array{Float64,1}},RealDomain{Float64,Int64,Float64}}}},getfield(Main, Symbol("##1257#1264")),getfield(Main, Symbol("##1258#1265")),getfield(Main, Symbol("##1259#1266")),ODEProblem{Nothing,Tuple{Nothing,Nothing},false,Nothing,ODEFunction{false,getfield(Main, Symbol("##1260#1267")),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{,Tuple{}}},DiffEqBase.StandardODEProblem},getfield(Main, Symbol("##1262#1269")),getfield(Main, Symbol("##1263#1270"))},Array{Subject{NamedTuple{(:dv,),Tuple{Array{Union{Missing, Float64},1}}},Nothing,Array{Pumas.Event{Float64,Float64,Float64,Float64,Float64,Float64,Int64},1},Array{Float64,1}},1},Optim.MultivariateOptimizationResults{Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat},Float64,Array{Float64,1},Float64,Float64,Array{Optim.OptimizationState{Float64,Optim.BFGS{LineSearches.InitialStatic{Float64},LineSearches.BackTracking{Float64,Int64},getfield(Pumas, Symbol("##220#221")){NLSolversBase.OnceDifferentiable{Float64,Array{Float64,1},Array{Float64,1}},Array{Float64,1}},Nothing,Optim.Flat}},1},Bool},Pumas.NaivePooled,Array{Array{Any,1},1},Tuple{},Base.Iterators.Pairs{Symbol,Tsit5,Tuple{Symbol},NamedTuple{(:alg,),Tuple{Tsit5}}},TransformVariables.TransformTuple{NamedTuple{(:θ, :σ\_prop),Tuple{Pumas.ElementArrayTransform{TransformVariables.ShiftedExp{true,Float64},1},TransformVariables.ScaledShiftedLogistic{Float64}}}}}) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\display.jl:96  
[14] #invokelatest#1 at .\essentials.jl:790 [inlined]  
[15] invokelatest at .\essentials.jl:789 [inlined]  
[16] execute\_request(::ZMQ.Socket, ::IJulia.Msg) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\execute\_request.jl:112  
[17] #invokelatest#1 at .\essentials.jl:790 [inlined]  
[18] invokelatest at .\essentials.jl:789 [inlined]  
[19] eventloop(::ZMQ.Socket) at C:\Users\awolf-yadlin.juliapro\JuliaPro\_v1.2.0-1\packages\IJulia\fRegO\src\eventloop.jl:8  
[20] (::getfield(IJulia, Symbol("##15#18")))() at .\task.jl:268

---

<div class="post-metadata">

### Author: ![andreasnoack](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/andreasnoack/32/22_2.png) [@andreasnoack](https://discourse.pumas.ai/u/andreasnoack)
#### Post date: [December 4, 2019, 8:02am UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/11 "2019-12-04T08:02:23Z")

</div>

The error you get is a bug in Pumas when a parameters vector has ten or more elements. However, it only affects the printing of the fitted model. You shouldn’t get the error if you add a semicolon at the end of the line where you fit the model. You should be able to extract the fitted coefficient with `coef(res)` where `res` is the fitted model.

---

<div class="post-metadata">

### Author: ![andreasnoack](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/andreasnoack/32/22_2.png) [@andreasnoack](https://discourse.pumas.ai/u/andreasnoack)
#### Post date: [December 4, 2019, 8:47am UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/12 "2019-12-04T08:47:33Z")

</div>

Issue filed in [https://github.com/PumasAI/Pumas.jl/issues/779](https://github.com/PumasAI/Pumas.jl/issues/779)

---

<div class="post-metadata">

### Author: ![pkofod](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pumas.ai/pkofod/32/59_2.png) [@pkofod](https://discourse.pumas.ai/u/pkofod)
#### Post date: [December 4, 2019, 10:35pm UTC](https://discourse.pumas.ai/t/estimate-with-missing-values-on-dependent-variable/96/13 "2019-12-04T22:35:03Z")

</div>

@alewolf Thanks for finding and reporting on this issue. It has been fixed on the master branch, but obviously won’t be available before the next version of Pumas. For now, I unfortunately have to ask you to suppress the output or split your variables into vectors of length up to 9. Happy to help if you have some more specific code to share (with details stripped that are not for public consumption).