CODEWITHSUNDEEP
rstatistics
J. Kang
J. Kang

Reputation: 83

R: Shapiro-Wilk Test

I am using the Shapiro-Wilk Test to reject the null hypothesis of Normality.

I am using code from Yahoo. For example, for S&P500:

SP500 = get.hist.quote(instrument = '^GSPC',
        start="2001-01-01",  end = "2015-12-31",
        quote = c("AdjClose"),provider = "yahoo",
        compress = "d") 
rSP500 = diff(log(SP500)

I get an error when I try to run the Shapiro-Wilk Test.

For example, 

> SWTestSP500 = shapiro.test(rSP500[(3000-499):3000])
Error in if (rng == 0) stop("all 'x' values are identical") : 
  argument is of length zero

The only search I got online was that it has to do with the input not being numerical, but I am fairly certain that rSP500 are all numbers.

Can someone advise? Thank you!

Upvotes: 1

Views: 1493

Answers (1)

J. Kang
J. Kang

Reputation: 83

After reading up on the R documentation many times, I tried

SWTestSP500 = shapiro.test(as.vector(rSP500[(3000-499):3000]))

And it finally worked. Turns out that shapiro.test() strictly takes in vectors.

Upvotes: 2

Related Questions