Registered User Joined: 2/15/2010 Posts: 6
|
Can you explain the PGR 1yr that is available in easy scans. The function ask for a value. What is the definition of this value.
|
Worden Trainer
Joined: 10/7/2004 Posts: 65,138
|
We calculate Price Growth Rate by creating a Linear Regression of the natural log of price over the specified timeframe. The slope of the Linear Regression is then compounded over a year and converted to a percentage.
-Bruce Personal Criteria Formulas TC2000 Support Articles
|
Registered User Joined: 6/19/2014 Posts: 40
|
Bruce:
I've been comparing PGR to actual growth & sometimes it's the same; sometimes it's much higher. Can you explain the usefulness please of PGR as an evaluation tool vs. absolute price growth, and can you also tell me how to calculate the actual 1 yr. price gain and YTD price gain, so I can add to a column?
Also, in Easy Scans, can you tell me how to filter out non-flagged symbols so only flagged show?
Lastly, I am not receiving replies to topics via email and wondered if I've missed a setting?
Thanks, Fred
|
Worden Trainer
Joined: 10/7/2004 Posts: 65,138
|
I can usually explain how something is done, but I can't actually make suggestions as to the why you or somebody else would want to do one thing instead of another thing. I'll try to give you some insight without actually crossing the line however.
Add a Linear Regression Line indicator to price. The details aren't that important, but what it does do is create a best fit line through the data. Prices of real live actively trading stocks are almost never going to line up exactly with this line. Some of the prices will be above the linear regression line and other prices will be below the linear regression line.
A Linear Regression Line places the greatest weight in the calculations of the slope of the line on the bars at the beginning and end of the Linear Regression Line and places the least weight in the calculations of the slopo ef the line on the bars in the middle of the Linear Regression Line. Despite this, it is is still using all of the data to the slope of the linear regression line and is not completely ignoring the data in the middle.
If you are just calculating a price percent change, it is only going to use the price of the first and last bars in its calculations. None of the date in the middle will be considered at all. If this is a good or bad thing for what you are trying to do is entirely up to you.
If I were to calculate a 1 year price percent change, I would probably use the following daily Indicator Formula.
100 * (C / C252 - 1)
Note that 252 bars ago is not going to generally fall on the same date last year as the current date, but 252 is very good approximation (people generally use a number of trading days from 250 to 260).
If I were to calculate a YTD price percent change, I would probably us ethe following yearly Indicator Formula.
100 * (C / C1 - 1)
Note that it has the same structure as the first formula. The reason it works is that in a 1 year time frame, C1 is the closing price of the previous calendar year and C is the current price.
You can sort flagged symbols to the top of an EasyScan, but the only way to have only flagged symbols show in the EasyScan results would be to use All Flagged Symbols as List to Scan of the EasyScan.
Building a Scan with Multiple Conditions (7:29)
To receive notification of replies to posts in the forums (you won't actually get the replies by email), you need to select Watch this topic at the top of the topic or select Watch Forum at the bottom of the forum in which the topic is posted (but this second option will give you notice of all posts in that forum).
Learn how to use the forums: post a new topic, reply, Search existing topics
Note that the email could also be sorted into a spam folder of some sort or get blocked by a spam filter. So if your email has a spam folder, you might want to check there.
-Bruce Personal Criteria Formulas TC2000 Support Articles
|