I am scanning using the following custom indicator for the previous quarter's percentage change:

100 * (C2 / C3 - 1)

And scanning with this custom indicator for the current quarter's percentage change:

100 * (C1 / C2 - 1)

As examples, a few stocks such as VERI, HCC, GDI, and BTU, are new to being publically-traded and thus have reported earnings in just the recent quarter.  I would like a way to separate those out from others that have reported earnings in the previous quarter.

I've tried various ways of doing this, but have been unsuccessful.

Can you come up with a way to scan for the previous quarter's percentage change being non-existent?  Or, can you suggest a way of changing the previous quarter's percentage change to equal zero?

Or, maybe there's another way to achieve my desired result.  I am open to whatever you can suggest.

It may just be the way we are thinking about current vs previous, but I would think of the current quarter as (this would be quarter to date):

100 * (C / C1 - 1)

And the previous quarter as:

100 * (C1 / C2 - 1)

You can create a sort for not having a price percent change in the previous, but not really a scan. If the formula can't be calculated, it will be effectively be treated as false, and the symbol will not show up in the scan.

You said, "You can create a sort for not having a price percent change in the previous, but not really a scan. If the formula can't be calculated, it will be effectively be treated as false, and the symbol will not show up in the scan."

Let's try this way, because I haven't been able to get either of the above formulas to work.  Maybe I need more direction on this.

As far as sorting goes, you can just sort using your formulas.

If they return a value, the symbols have enough data. If the WatchList Column displays ... instead, then there isn't enough data to calculate the formula (the symbol is too new).

For the first formula 100 * (C1 / C2 - 1) it does return a negative value (but not zero) for three of the stocks mentioned: VERI, BTU, GRI.  For the last stock, HCC, it returns a positive value.

Your suggestion about sorting is difficult when I am scanning 7,204 US Common Stocks, ADRs, and ETFs. 

That is why I am trying to find another way to efficiently scan for non-existent earnings reported in the most previous quarter and earning reported in the most current quarter -- or for a zero percentage change in the most previous quarter, with a positive or negative percentage change in the current quarter.

I've tried doing this through fundamentals: "EPS Percentage Change Latest Qtr" and "EPS Percentage Change 2nd Qtr Back".

I've also tried your percentage formulas.

What else can we try?

Well they were around long enough that they could have reported earnings. That said, try this.

Sort the list and look for ... instead of values (0.00) doesn't count, it just means there wasn't a change).

Note that VERI, BTU, and HCC all seem to have reported at least two quarters of earnings because there are at least two different earnings values.

If you want to go back another quarter, change the ROC% period from 1 to 2 and add it as a WatchList Column.

If sorting isn't going to work for you for this in TC2000 then there is isn't any else for you to try beyond that in TC2000.

You've said, "Note that VERI, BTU, and HCC all seem to have reported at least two quarters of earnings because there are at least two different earnings values."

To check this premise, I've just added EPS for each of the last four quarters as columns.  The only one that has reported earnings for each of those four quarters is BTU.

The three others -- HCC, GDI, and VERI -- show no reported earnings for any of the four quarters, even for the most recent quarter.

I think the problem may be that the mathematical formulas I am using for each of the last four quarters do not give entirely accurate values because these general formulas do not account for weekend days or stock market holidays.

Is my thinking correct?

We don't have EPS columns available for anything besides the current quarter. It is called EPS Latest Qtr.

The other columns are EPS Percent Change columns and compare that quarter to comparible quarter for the previous year. So 2nd quarter back would need at least 5 quarters of earnings report to calculate, not 2.

Please try adding a EPS Latest Qtr to the chart and looking at the plot. If it has gone up and down, then there has to be more than one earnings report.

Doing as you have suggested is interesting.  On a 1Y chart, it shows the latest quarter of earnings for BTU, and no earnings for the other three stocks: GDI, HCC, VERI.  More in a minute...

You will be able to see on the chart where the EPS Latest Qtr changes if you switch to a daily time frame.

The suggestion of using a quarterly time frame was specifically for the purpose of creating the ROC% indicator to use as a WatchList Column and sort.

"Next Earnings" column from TC agrees with Earnings Whispers that next earnings will be on January 24, 2018 (a Wednesday).  Thus, perhaps I can assume that the most recent earnings report was on or about October 24, 2017 (a Tuesday).

What puzzles me now is that the "EPS Percent Change Latest Qtr" plotted on my chart shows a beginning on either 9/29/17 extending through to today (and probably beyond, I assume).

Can you explain this?

Does the quarter end on or about 9/29/17, but the earnings are reporter later, on or about 10/24/17?

There is an extensive discussion about why the EPS Latest Qtr values change on the dates they do in the EPS, Fundamental Data topic. StockGuy's posts explore why  But the dates relate to actual end of the fiscal quarter for each company and not the date of the earnings announcement.

I will try your suggestions as repeated below and get back to you with any questions or comments later.

Sort the list and look for ... instead of values (0.00) doesn't count, it just means there wasn't a change).

Note that VERI, BTU, and HCC all seem to have reported at least two quarters of earnings because there are at least two different earnings values.

If you want to go back another quarter, change the ROC% period from 1 to 2 and add it as a WatchList Column.

I just completed your steps two ways.

First, I added Rate of Change % as a separate plot under EPS Percent Change Latest Qtr,  Changing the period to 1 and selecting as "Data Source"  EPS Percent Change Latest Qtr, then finally copying as a column, these were the results:

ROC % 1 Quarterly

BTU 0.00

Others ---

Second, I added Rate of Change % as a child plot of EPS Percent Change Latest Qtr and completed the other steps, and this gave the same results as above.

Both are different than what you said.

I am wondering why?

The current quarter ends December 31, 2017 (but the last trading day will be December 29, 2017).

The previous quarter ended September 30, 2017 (but the last trading day was September 29, 2017).

Two quarters ago ended June 30, 2017.

Three quarters ago ended March 31, 2017.

Four quarters ago ended December 31, 2017 (but the last trading day was December 30, 2017).

So the values of EPS Latest Qtr are going to be compared on those particular dates. We are using a quarterly time frame because it is automatically going to get values exactly one quarter apart, but that is really only true for previous quarters since the current quarter has not ended.

I never said what the values would be for particular symbols. Only that you were sorting to eliminate items with ... as the result because they were too new to calculate a rate of change in the EPS Latest Qtr.

With a 1 period rate of change, this would happen if there was no value at the end of previous quarter. So there must be a fiscal quarter ending on or or before September 29, 2017 with EPS reported if anything besides ... is returned.

With a 2 period rate of change, this would happen if there was no value at the end of two quarters ago. So there must be a fiscal quarter ending on or before June 30, 2017 with EPS reported if anything besides ... is returned.

With a 3 period rate of change, this would happen if there was no value at the end of three quarters ago, so there must be a fiscal quarter ending on or before March 31, 2017 with EPS reported if anything besides ... is returned.

With a 4 period rate of change, this would happen if there was no value at the end of four quarters ago, so there must be a fiscal quarter ending on or before December 30, 2017 with EPS reported if anything besides ... is returned.

The most recent quarter is shorter than the other quarters and this is going to affect your count of the number of EPS reports. I guess you can try to use rates of change in multiples of 13 weeks or 63 days, but I am really not sure it is going to be any better for what you are trying to do.

Okay.  So you are saying that Worden's Fundamental date (EPS Percent Changes) are based on calendar quarters. (Jan-Mar, Apr-June, July-Sept, and Oct-Dec)?

However, these quarters do not coordinate with the reporting dates used by individual publically-traded companies?

Yes, the earnings data and dates are based on the fiscal quarters for the company. Those fiscal quarters will necessarily line up exactly with calendar quarters. A quarterly time frame will line up exactly with calendar quarters.

Earnings are generally going to be reported after the end of the fiscal quarter ends because it takes time for a company to properly compile and report earnings data for the quarter.

The data will line up with the end of the fiscal quarter and not with the earnings report.

Thanks for your suggestion to read the discussion regarding reporting of EPS between your Admin person, StockGuy, and long-time subscriber, ThnkBigr, on the site below:

http://forums.worden.com/default.aspx?g=posts&t=67395

My experience using a "Next Earnings" column offered by Worden has been frustrating.  I've found too many inaccuracies to use it, so I've had to resort to using a service called "Earnings Whispers" whose dates are accurate, and in addition, Earnings Whispers reports whether earnings are reported before the market open -- or after the market close.

I assume Worden is still using Morningstar for its' data.  Either their data is not totally accurate -- or as ThnkBigr pointed out, there is a problem with Worden's updating of data regarding weekends.  (I am not quite clear how this works).

The most important point made by ThnkBigr is that earnings reports should show up not at the close of the quarter, but on the reporting date, both in columns and on the chart.

I hope you will pass my opinion on to the power that be to improve TC2000 for active traders.

Just one more comment.  It seems Worden is making it easier for active traders with its' Worden Brokerage and chart capabilities to allow trading on the same screen as analysis, not only for stocks but for options. 

It would seem that this is the time to add more tools that active traders need such as the accurate next earnings reporting dates -- both before market open and after market open -- that would provide additional tools for active traders.

Also with date formats allowing for easy sorting: YYMMDD

Thank you for your suggestion.