Well, now after trying again, it works properly.
I have no idea what have changed, but you may see yourself from my earlier example that the workfile was undated and consequently TA moving averages were from the future.
Let's see if this issue might occur one more time
Search found 13 matches
- Tue Aug 07, 2012 1:20 pm
- Forum: Add-in Support
- Topic: GetStocks (fetch stock data)
- Replies: 52
- Views: 54581
- Tue Aug 07, 2012 3:14 am
- Forum: Add-in Support
- Topic: TechAsis (Technical Analysis)
- Replies: 1
- Views: 10598
Re: TechAsis (Technical Analysis)
Bollinger Bands: 1. It seems that the default series OUT_HIGH should be renamed OUT_LOW as it provides values for lower boundary (the same for OUT_LOW, change the name to OUT_HIGH) 2. The resulting default graph OUT_G that appears at the end of analysis would become more meaningful if you add the se...
- Tue Aug 07, 2012 2:54 am
- Forum: Add-in Support
- Topic: GetStocks (fetch stock data)
- Replies: 52
- Views: 54581
Re: GetStocks (fetch stock data)
Is it possible to download daily stock data into dated workfile? Weekly and monthly data are OK, but dailies are totally messed up: when the data are downloaded into undated workfiles most recent quotes appear at the top of the data sheet and consequently moving averages are calculated forward looki...
- Mon Oct 25, 2010 1:34 am
- Forum: Econometric Discussions
- Topic: mean reversion speed
- Replies: 6
- Views: 13321
Re: mean reversion speed
Sorry, you're right, for autoregressive AR(1) process presented in first post the formula is ln(0.5)/ln(b).
Mine is also correct I guess, but for Ornstein–Uhlenbeck representation.
Mine is also correct I guess, but for Ornstein–Uhlenbeck representation.
- Sun Oct 24, 2010 9:57 am
- Forum: Econometric Discussions
- Topic: regression in engle and granger cointegration
- Replies: 1
- Views: 3010
Re: regression in engle and granger cointegration
Though this may sound counter-intuitive, it does matter which one of the two -- or three, or five -- variables you choose as a dependent variable. In general, you will follow the stats from Engle/Granger single equation test and choose the variable that is strongest in rejecting H0 (in humane langua...
- Sun Oct 24, 2010 5:39 am
- Forum: Econometric Discussions
- Topic: mean reversion speed
- Replies: 6
- Views: 13321
Re: mean reversion speed
This is a little bit tricky. Mean reverting value can only be reached in the limit (as the forecast horizon approaches to infinity). Therefore, "half-life" is the most common way of measuring the speed of mean reversion. This is to calculate the number of periods needed for the forecast t...
- Thu Oct 14, 2010 12:38 am
- Forum: Programming
- Topic: NA in "if" statement
- Replies: 4
- Views: 4991
Re: NA in "if" statement
Thank you for the help
- Wed Oct 13, 2010 11:50 pm
- Forum: Programming
- Topic: NA in "if" statement
- Replies: 4
- Views: 4991
Re: NA in "if" statement
@recode will work for the first time, but when I use it for the second, it will overwrite the first results.
Any other suggestions?
Any other suggestions?
- Wed Oct 13, 2010 12:59 pm
- Forum: Programming
- Topic: NA in "if" statement
- Replies: 4
- Views: 4991
NA in "if" statement
I'm trying to assign +1 values to a 'trade' series when my 'res' (=residual) series is greater than standard deviation (=s) and -1 when my 'res' series less than negative standard deviation if res>s then series trade=1 else if res<-s then series trade=-1 endif endif but getting a message "NA fo...
- Mon Oct 11, 2010 11:49 am
- Forum: Bug Reports
- Topic: Variance ratio test
- Replies: 2
- Views: 5518
Re: Variance ratio test
Got it.
Quite interesting.
Thanks.
Quite interesting.
Thanks.
- Sun Oct 10, 2010 4:25 am
- Forum: Bug Reports
- Topic: Variance ratio test
- Replies: 2
- Views: 5518
Variance ratio test
I'm running variance ratio test on three time series: 1. lkoh, original time series of LKOH stock prices. Data specification --> exponential random walk (the rest is default) 2. ll, which is log(lkoh). Data specification --> random walk 3. dl, which is dlog(lkoh). Data specification --> random walk ...
- Tue Oct 05, 2010 4:48 am
- Forum: Bug Reports
- Topic: "Fill" bug?
- Replies: 1
- Views: 3194
"Fill" bug?
When trying to fill matrix by rows, any attempt to fill any row other than the first one encounters an error message: "Incorrect number of elements provided for fill" A simple code generating the error looks like this: create u 100 matrix(10,10) y y.fill(o=11,b=r) 1,1 There is no error, if...
- Thu Aug 19, 2010 1:48 pm
- Forum: Programming
- Topic: Rolling cointegration
- Replies: 1
- Views: 4125
Rolling cointegration
Hello,
Is there a way to save results, i.e. tau statistic or p-values, from single equation cointegration tests like Engle-Granger, Phillips-Ouliaris etc?
The question in particular concerns performing and saving results from a rolling cointegration test.
Your answer is really appreciated.
Is there a way to save results, i.e. tau statistic or p-values, from single equation cointegration tests like Engle-Granger, Phillips-Ouliaris etc?
The question in particular concerns performing and saving results from a rolling cointegration test.
Your answer is really appreciated.