SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Strategies & Market Trends : Technical Analysis- Indicators & Systems
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Bill Sandusky who wrote (2350)8/3/1997 12:23:00 AM
From: GW Riedeman   of 3325
 
Bill...That's interesting....A linear regression line calculated normally would be curved downward when the same data is plotted on semilog scale. "Linear" regression means just that, "linear." A "semilog" plot is NOT linear.

For comparison:

A straight, inclined upward line on a LINEAR plot will curve DOWN at higher "X" values on the corresponding SEMILOG plot, approaching horizontal at larger numbers.

A straight, inclined upward line on a SEMILOG plot will curve UP sharply at higher "X" values on the corresponding LINEAR plot, eventually pointing almost straight up.

By "corresponding," I mean the same X-Y values plotted on each chart.

If you get a STRAIGHT line on TradeStation when creating a linear regression on BOTH the LINEAR and SEMILOG.plots, Trade Station is doing something they ought to be very clear about in their instructions: They are fooling with your data! This isn't bad, but they ought to let you know about it.

The algegra is pretty simple:

The equation for the LINEAR least squares line is (plots as a straight line on a LINEAR graph):

$$$ = (slope) x (periods) + (price at period 0)

The equation for the SEMILOG least squares line is (plots as a straight line on a SEMILOG graph):

Log($$$) = (slope) x (periods) + (Log{price} at period 0)

I'm not surprised that there are minor differences in the indicator crossings; Most charting programs use smoothing algorithms to make a nice looking graph out of the discrete data points, and they often move the data around a little bit for aesthetic purposes. Always be a little cautions when drawing conclusions from smoothed data, particularly when you are making a close call. Go the the actual numbers most programs will show for your cross hairs or other cursor.

I hope this helps. Any more questions, just ask...Bill
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  

HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2025 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News