Different p-value compare to R

[sourish-cmi]

Hi

In R, I did the Whites' test (aka. Brush Pegan Test). I found the p-value is different. Can you please have a look and check if there is a mistake in my code.

The R-code and output

R> attach(mtcars)
R> model = lm(mpg~hp+wt+gear,mtcars)
R> library(lmtest)
R> bptest(model,data=mtcars)

	studentized Breusch-Pagan test

data:  model
BP = 1.3627, df = 3, p-value = 0.7143

The Julia code and output

using HypothesisTests

function BPTest(container:: FrequentistRegression, data::DataFrame) 
       fm_frame = ModelFrame(container.formula, data);
       X=modelmatrix(fm_frame)
       e = residuals(container)
       BP_Test1 = WhiteTest(X,e);
       return(BP_Test1);
end
BP_tst = BPTest(mod1,train_df)

White's (or Breusch-Pagan's) test for heteroskedasticity
--------------------------------------------------------
Population details:
    parameter of interest:   T*R2
    value under h_0:         0
    point estimate:          8.73554

Test summary:
    outcome with 95% confidence: fail to reject h_0
    one-sided p-value:           0.4620

Details:
    T*R^2 statistic:        8.73554
    degrees of freedom:     9
    type:                   White

pvalue(BP_tst)
0.46203539153321815