[R-SIG-Finance] Run tests Fama

[Patrick Burns]

I would think that any difference in the details
of the definition would be inconsequential.

Something like:

rle(sign(ret))

should do.


Patrick Burns
patrick at burns-stat.com
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")

Bastian Offermann wrote:
> Hello list,
>
> is anyone of you familiar with Fama's run tests in his 1965 behaviour 
> of stock market prices dissertation? I am not quite sure if I 
> understand the concept of "runs" correctly. He says *"A run is defined 
> as a sequence of price changes of the same sign. For example, a plus 
> run of length i is a sequence of i consecutive positive price changes 
> preceded and followed by either negative or zero changes."* So far so 
> good.
>
> -1  +5  +6  + 7 -2
>
> So, 5,6,7 should be a plus run of length 3. The total number of runs 
> is thus 1 plus, 0 minus, 0 unchanged.
>
> From a different source I get the impression that the runs are counted 
> differently. E.g.
>
> 0,-10,1,1,-12,1,-10,-5,1,4,-4,1,0,-11,3,-12,-10,-1,0,1
>
> 13 runs. It does not really say how many of each, confusing me.
>
> Any my code does not really match the result for the long vector. 
> Thanks for your input.
>
> Regards
>
>
>
> rsp500 <- c(0,-10,1,1,-12,1,-10,-5,1,4,-4,1,0,-11,3,-12,-10,-1,0,1)
> rsp500 <- c(-1,5,6,7,-2)
> pos <- rsp500[rsp500>0]
> neg <- rsp500[rsp500<0]
> m <- (length(rsp500)^2 + length(rsp500) - length(pos)^2 - 
> length(neg)^2)/length(rsp500) # no. of expected runs
> runplus <- 0
> runminus <- 0
> runequal <- 0
> i <- 1
> while(i <(length(rsp500)-1)) {
>                            if(rsp500[i+1] > 0 & rsp500[i] < 0) runplus 
> <- runplus + 1
>                            if(rsp500[i+1] < 0 & rsp500[i] > 0) 
> runminus <- runminus + 1
>                            if(rsp500[i+1] < 0 & rsp500[i] < 0 & 
> rsp500[i+1] == rsp500[i]) runequal <- runequal + 1
>                            if(rsp500[i+1] > 0 & rsp500[i] > 0 & 
> rsp500[i+1] == rsp500[i]) runequal <- runequal + 1
>                            i <- i + 1                          
>                           }
>
> _______________________________________________
> R-SIG-Finance at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only.
> -- If you want to post, subscribe first.
>
>