| Index | Recent Threads | Who's Online | User List | Search |
|
|
 |
Portfolio123 » List all forums » Forum: Feature Suggestions » Thread: Feature Request: Standard Deviation of Prices |
|
Total posts in this thread: 3 |
[Request a Feature] [Add To My Favorites] [Watch this Thread] |
| Author |
|
|
rcflyerco
Advanced Member
|
Standard Deviation of the closing prices could be used to calculate Bollinger bandwidths and volatility in general. I believe this is distinct from the PctDev function currently available, although there may be a way to estimate standard deviation based on the percent deviation of the percentage price moves. (Help me if I am missing something obvious here.) The most useful parameter would be the number of periods over which to calculate the standard deviation. Additional useful parameters would be: Number of bars per period. Whether to use close, open, hi, or low prices for the calculation. Offset. Offset would allow comparing the 20 bar volatility last month to the 20 bar volatility this month, for example. Alternatively, a function which returns price as a percentage of the Bollinger bandwidth would be helpful. A similar request was discussed here - http://www.portfolio123.com/mvnforum/viewthread?thread=1013#3374 View the feature request here. |
||
|
|
Sterling
Advanced Member
|
I was asking something similar previously and after the updated correction and looking at the detailed function reference here I get the feeling one should be able to simulate a Bollinger band from the PctDev function. I haven't tried implementing it yet however and haven't looked at it too closely but my interpretation is that PctDev returns 1 standard deviation expressed as a percentage: PctDev = 1 standard deviation (expressed in %, not standard deviation). e.g. If PctDev = 6.23, 6.23% equals one standard deviation (as opposed to 6.23% of one standard deviation). Therefore to calculate 20-period 2 standard deviation bollingers bands : Upper Bollinger Band = MA(20) * (1 + (2*PctDev(20,1)/100) Lower Bollinger Band = MA(20) * (1 - (2*PctDev(20,1)/100) ( I am multiplying the MA by 100% + 2*PctDev and 100% - 2*PctDev) Again this is based on an on the spot interpretation. I could be completely wrong. Please double check and verify to see if it makes sense and correct me if I'm wrong. Rcflyerco I think you and I may be a little disoriented coming at it after having used charting applications which made it so easy to apply standard deviation to prices in terms of standard deviation. Excel users I suspect apprehend the presentation of PctDev in terms of % more quickly. Anyway the above does seem a somewhat convoluted way of doing things and a specialized function would simplify and prove more useful. As measures of volatility go though I think I'd support more the development of ATR. ---------------------------------------- [Edit 3 times, last edit by Sterling at Sep 1, 2006 11:48:06 PM] |
||
|
|
rcflyerco
Advanced Member
|
Closed by author |
||
|
| [Show Thread Printable Version] |
