New Members: Be sure to confirm your email address by clicking on the link that was sent to your email inbox. You will not be able to post messages until you click that link.

SCTR ratings

isabelds
isabelds ✭
in Scanning
This is a two fold question.

First, why are only certain stocks rated with SCTR? For example, two coal stocks, sxc and nrp.

Second, what would be an alternative rating system/indicator I can scan with that might bring me similar results to SCTR?

Thank you in advance!

Comments

  • markd
    markd mod
    To your first question, here's an excerpt from the "Scanning Technically" blog from a while ago that explains it:

    "Initially there are five SCTRs covering the S&P 500, S&P 400, S&P 600, US ETFs and stocks listed on the Toronto exchange. Currently no stock will have more than one SCTR score as the above group memberships happen to be mutually exclusive.

    This will likely change as more SCTR groups are created in the future. Thus only stocks in these 5 groups will have SCTR values initially. Stocks outside these groups will have no SCTR value.

    In the case of the ETFs and Toronto exchanges, we're choosing only stocks that trade above $1 and have volume greater than 40,000 shares using a 20 day simple moving average."

    London and India have been added since. Also, Russell 2000 stocks have been added, so the small cap rankings now include more than just the SP600. It's possible there have been other changes, but I'm not aware.

    To your second question, SCTR rankings include performance in several time frames, so its hard to replicate with just one indicator. But, a rough way to do it might be to rank your scan results by ROC(?). So, for instance, to get the best performers over the last year you might end your scan with

    rank by ROC(251)

    Over the last quarter:

    rank by ROC(63)

    Over the last month:

    rank by ROC(21)

    If you have a very long results lists, you might first rank by annual performance, and delete the bottom half of the list, then rank by quarterly performance and delete the bottom half of that list, and finally rank that list by most recent monthly performance. Or, you could start with best most recent monthly performance, then quarterly then annual. Experiment and see.
  • markd
    markd mod
    This is another possibility - you could weight the Fast Stochastics values of different time frames - or maybe you could do the same with ROC

    rank by Fast Stoch %K(251,1) * 4 + Fast Stoch %K(63,1) * 3 + Fast Stoch %K(21,1) * 2 + Fast Stoch %K(10,1)

    I am checking with support, but this statement should do the multiplication first, then add the products. If you use brackets to make sure, you don't get valid results.

    You can play around with the weightings - e.g. change 4 to 2, 3 to 4, 2 to 5, or whatever, to see how it affects the results.

    As it is, it ranks things somewhat like SCTR, in that higher SCTR scores are generally nearer the top and an lower ones near the bottom, but it doesn't really mimic it VERY closely.
  • isabelds
    isabelds ✭
    Thank you for your thoughts on this, I am going to experiment with ROC and the stochastics.
  • markd
    markd mod
    edited November 2016
    Hi @isabelds , the rank by statement with stochastics may not work. It passes syntax and gives results, but I'm not sure the results make sense - I don't know how the scan engine is making the calculation, so I don't know if the results are meaningful. Support tells me the rank by statement functionality is pretty basic - it's only meant to handle one or two simple operations, not a complex one like this. So, the ROC approach should work. The Fast K approach is questionable.

Categories