Message Boards Message Boards

Calcuate the error between two histograms and also p-value?

Posted 3 years ago

Dear all,

Is it possible calculate the error between the two smooth histograms and p-value for the model values?

data = {1.2803`, 1.3201`, 1.3293`, 1.3218`, 1.3083`, 1.3173`, 1.3199`,
    1.3199`, 1.3003`, 1.302`, 1.2893`, 1.3005`, 1.3155`, 1.3138`, 
   1.3186`, 1.3194`, 1.3127`, 1.3114`, 1.3335`, 1.3437`};

model = {1.30276835568204`, 1.28597164114136`, 1.30781887157607`, 
   1.32399857112606`, 1.3125415849624`, 1.33466340860532`, 
   1.34154207756901`, 1.33888043871135`, 1.29826352017477`, 
   1.29423486081725`, 1.29031985739288`, 1.29343024958048`, 
   1.29583849802769`, 1.28302041907369`, 1.25946083824112`, 
   1.26819316624496`, 1.2553816286335`, 1.26094249800321`, 
   1.22202625825362`, 1.20903091640703`};

SmoothHistogram[{data, model}, 
 PlotStyle -> {{Dashed, Thickness[.004]}, {Orange, Thickness[.002]}}, 
 LabelStyle -> {17, GrayLevel[0]}, ImageSize -> 700, PlotRange -> All,
  PlotLegends -> {"data", "model"}, 
 AxesLabel -> {"EUR/USD", "SmoothHistogram"}]

enter image description here

POSTED BY: Alex Teymouri
3 Replies
Posted 3 years ago

You should wean yourself from relying solely on P-values. See the statement from the American Statistical Association.

POSTED BY: Jim Baldwin
Posted 3 years ago

What do you mean by "error" ? Maximum difference between the CDF's? Area between the PDF's? Difference in means? You might want to ask this question first on CrossValidated and then back here for implementation in Mathematica.

POSTED BY: Jim Baldwin
Posted 3 years ago

Dear Jim,

I meant to calculate the area difference between the two curves as below: Error= Area der model -Area under data Fortunately, I found the solution.Thank you .

If possible, please let me know, it is possible to estimate the p-value?

POSTED BY: Alex Teymouri
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract