CODEWITHSUNDEEP
r
Judy S
Judy S

Reputation: 145

Error code 100 fitting exp distribution using fitdist in r

I am trying to fit an exponential distribution to my data but I get the error below

"Error in fitdist(x41, "exp", method = "mle") : 
  the function mle failed to estimate the parameters, 
                with the error code 100"

I have tried mme which works but I have other distributions with mle so I need mle in exponential distribution as well. I have been stuck for days. Can anyone help me, please?

My data looks like this.

2795.5
304.6833
2786.45
5897.75
4381.367
1178.1
351.3167
109.85
459.6167
13.26667
0.033333
846.3833
3698.45
1527.1
94.31667
15.01667
271.8833
473

This is my code

ExpMle41 <- fitdist(x41, "exp", method="mle") 
ExpMle41
plot(ExpMle41)

Any help will be greatly appreciated. Thank you.

Upvotes: 13

Views: 20793

Answers (1)

Spacedman
Spacedman

Reputation: 94182

Assuming this is fitdist from fitdistrplus package, I can duplicate your error:

> fitdist(x41, "exp", method="mle")
<simpleError in optim(par = vstart, fn = fnobj, fix.arg = fix.arg, obs = data,     gr = gradient, ddistnam = ddistname, hessian = TRUE, method = meth,     lower = lower, upper = upper, ...): non-finite finite-difference value [1]>
Error in fitdist(x41, "exp", method = "mle") : 
  the function mle failed to estimate the parameters, 
                with the error code 100

but there's some large numbers in your data... maybe if we scale it all down by a factor...

> fitdist(x41/10000, "exp", method="mle")
Fitting of the distribution ' exp ' by maximum likelihood 
Parameters:
     estimate Std. Error
rate   7.1417   1.683315

Well that seemed to work. Let's scale by a bit less:

> fitdist(x41/1000, "exp", method="mle")
Fitting of the distribution ' exp ' by maximum likelihood 
Parameters:
     estimate Std. Error
rate  0.71417  0.1683312

Right. Divide by a thousand works. Let's keep going:

> fitdist(x41/100, "exp", method="mle")
Fitting of the distribution ' exp ' by maximum likelihood 
Parameters:
     estimate Std. Error
rate 0.071417 0.01682985

Fine.

> fitdist(x41/10, "exp", method="mle")
Fitting of the distribution ' exp ' by maximum likelihood 
Parameters:
      estimate  Std. Error
rate 0.0071417 0.001649523

So scaling the data by 1/10 works, and you can see how the estimate and SE scale. Let's go one more step:

> fitdist(x41/1, "exp", method="mle")
<simpleError in optim(par = vstart, fn = fnobj, fix.arg = fix.arg, obs = data,     gr = gradient, ddistnam = ddistname, hessian = TRUE, method = meth,     lower = lower, upper = upper, ...): non-finite finite-difference value [1]>
Error in fitdist(x41/1, "exp", method = "mle") : 
  the function mle failed to estimate the parameters, 
                with the error code 100

Crunch. It looks like some numerical stability problem with the underlying algorithm. If its taking exponentials of your data at any point then maybe it hits something indistinguishable from infinity. Like:

> exp(x41)
 [1]           Inf 2.100274e+132           Inf           Inf           Inf
 [6]           Inf 3.757545e+152  5.096228e+47 4.064401e+199  5.776191e+05
[11]  1.033895e+00           Inf           Inf           Inf  9.145540e+40
[16]  3.323969e+06 1.195135e+118 2.638092e+205

But scale by ten and the maths can cope, just about (E+256!!!)

> exp(x41/10)
 [1] 2.552833e+121  1.706977e+13 1.032728e+121 1.367817e+256 1.907002e+190
 [6]  1.459597e+51  1.809216e+15  5.898273e+04  9.139021e+19  3.768462e+00
[11]  1.003339e+00  5.727429e+36 4.184491e+160  2.094645e+66  1.247731e+04
[16]  4.489166e+00  6.423056e+11  3.484408e+20

Upvotes: 19

Related Questions