# NA loo result

**URL:** <https://discourse.mc-stan.org/t/na-loo-result/7071>\
**Category:** Modeling\
**Tags:** loo\
**Created:** [December 24, 2018, 2:31pm UTC](https://discourse.mc-stan.org/t/na-loo-result/7071 "2018-12-24T14:31:45Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Tairan](https://avatars.discourse-cdn.com/v4/letter/t/34f0e0/32.png) [@Tairan](https://discourse.mc-stan.org/u/Tairan)\
**Post date:** [December 24, 2018, 2:31pm UTC](https://discourse.mc-stan.org/t/na-loo-result/7071/1 "2018-12-24T14:31:45Z")

</div>

Hello,

I have a question about loo and waic result. In my simulation study, I can receive loo and waic result within a reasonable range. However, in the real data, I got pretty huge waic result:

```
> result$waic_tweedie
$`estimates`
               Estimate SE
elpd_waic -2.090346e+65 1.04399e+65
p_waic 2.090346e+65 1.04399e+65
waic 4.180692e+65 2.08798e+65

$pointwise
            elpd_waic p_waic waic
   [1,] -1.945756e+22 1.945756e+22 3.891511e+22
   [2,] -3.020967e+20 3.020967e+20 6.041934e+20
   [3,] -6.054780e+19 6.054780e+19 1.210956e+20
   [4,] -1.172806e+48 1.172806e+48 2.345612e+48
   [5,] -1.252922e+28 1.252922e+28 2.505843e+28
   [6,] -9.437428e+24 9.437428e+24 1.887486e+25
   [7,] -4.874609e+47 4.874609e+47 9.749217e+47
   [8,] -4.280829e+20 4.280829e+20 8.561658e+20
   [9,] -1.071315e+23 1.071315e+23 2.142630e+23
  [10,] -8.666166e+47 8.666166e+47 1.733233e+48
  
 [313,] -4.071934e+27 4.071934e+27 8.143867e+27
 [314,] -1.524270e+27 1.524270e+27 3.048540e+27
 [315,] -3.188398e+42 3.188398e+42 6.376796e+42
 [316,] -2.061892e+30 2.061892e+30 4.123785e+30
 [317,] -5.827317e+26 5.827317e+26 1.165463e+27
 [318,] -2.773955e+20 2.773955e+20 5.547911e+20
 [319,] -3.483425e+42 3.483425e+42 6.966851e+42
 [320,] -2.785318e+31 2.785318e+31 5.570636e+31
 [321,] -5.451254e+40 5.451254e+40 1.090251e+41
 [322,] -2.305406e+45 2.305406e+45 4.610812e+45
 [323,] -1.153934e+22 1.153934e+22 2.307867e+22
 [324,] -9.492203e+45 9.492203e+45 1.898441e+46
 [325,] -9.666653e+32 9.666653e+32 1.933331e+33
 [326,] -2.442646e+36 2.442646e+36 4.885292e+36
 [327,] -6.533043e+43 6.533043e+43 1.306609e+44
 [328,] -3.177141e+23 3.177141e+23 6.354283e+23
 [329,] -3.417232e+24 3.417232e+24 6.834464e+24
 [330,] -8.447091e+28 8.447091e+28 1.689418e+29
 [331,] -7.124550e+19 7.124550e+19 1.424910e+20
 [332,] -6.053197e+19 6.053197e+19 1.210639e+20
 [333,] -3.058053e+54 3.058053e+54 6.116106e+54
 [reached getOption("max.print") -- omitted 1272 rows]

$elpd_waic
[1] -2.090346e+65

$p_waic
[1] 2.090346e+65

$waic
[1] 4.180692e+65

$se_elpd_waic
[1] 1.04399e+65

$se_p_waic
[1] 1.04399e+65

$se_waic
[1] 2.08798e+65

attr(,"dims")
[1] 12000 1605
attr(,"class")
[1] "waic" "loo" 

```

And the result for loo is even worse, with all NA values:

```
> result$loo_tweedie
$`estimates`
         Estimate SE
elpd_loo NA NA
p_loo NA NA
looic NA NA

$pointwise
        elpd_loo mcse_elpd_loo p_loo looic
   [1,] NA NA NA NA
   [2,] NA NA NA NA
   [3,] NA NA NA NA
   [4,] NA NA NA NA
   [5,] NA NA NA NA
   [6,] NA NA NA NA
   [7,] NA NA NA NA
   [8,] NA NA NA NA
   [9,] NA NA NA NA
  [10,] NA NA NA NA
  [11,] NA NA NA NA
  [12,] NA NA NA NA
  [13,] NA NA NA NA

$diagnostics
$diagnostics$`pareto_k`
   [1] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  [13] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  [25] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  [37] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  [49] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  [61] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  [73] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  [85] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  [97] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
 [109] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
 [121] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
 [133] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
 [145] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
 [157] Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf

$diagnostics$n_eff
   [1] NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA
  [17] NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA
  [33] NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA
  [49] NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA
  [65] NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA
  [81] NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA

$psis_object
NULL

$elpd_loo
[1] NA

$p_loo
[1] NA

$looic
[1] NA

$se_elpd_loo
[1] NA

$se_p_loo
[1] NA

$se_looic
[1] NA

attr(,"dims")
[1] 12000 1605
attr(,"class")
[1] "psis_loo" "loo"     

```

Do you have any idea where I did wrong? Thank you very much for your help.

I hope you all have a wonderful holiday season.

Best,

T.Y.

---

<div class="post-metadata">

**Author:** ![avehtari](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/avehtari/32/5935_2.png) [@avehtari](https://discourse.mc-stan.org/u/avehtari)\
**Post date:** [December 27, 2018, 3:04pm UTC](https://discourse.mc-stan.org/t/na-loo-result/7071/2 "2018-12-27T15:04:32Z")

</div>

Can you post your code? How did you check that loo is reasonable in the simulation study?