0

I am doing an lm()regression with R where I use stock quotations. I used exponential weights for the regression : the older the data, the less weight. My weights formula is like this : alpha^(seq(685,1,by=-1))) (the data length is 685), and to find alpha I tried every value between 0.9 and 1.1 with a step of 0.0001 and I chose the alpha which minimizes the difference between the predicted values and the real values. This alpha is equal to 0.9992 so I would like to know if it is statistically different from 1.

In other words I would like to know if the weights are different from 1. Is it possible to achieve that and if so, how could I do this ?

I don't really know whether this question should be asked on stats.stackexchange but it involves Rso I hope it is not misplaced.

etienne
  • 3,648
  • 4
  • 23
  • 37
  • 1
    It sounds like you're asking about appropriate hypothesis tests for your modeling setup, which would belong on stats.stackexchange.com instead of Stack Overflow. Stack Overflow would be best used if you knew what test you wanted to run but were having trouble with the implementation. – josliber Oct 09 '15 at 16:28
  • 1
    It should be in CV. But since you're here: you're also doing a `lm` on a model with correlated errors, which is inappropriate. – alexwhitworth Oct 09 '15 at 16:36
  • @Alex: why should I not use `lm`and what should I use instead ? – etienne Oct 09 '15 at 16:37
  • 2
    Because the assumptions of the linear model are violated with TS data. – alexwhitworth Oct 09 '15 at 16:38

0 Answers0