Several tweaks to our ratio engine, and a new Share function

Dear All,

We made the following tweaks to our engine:

  1. New function for non-diluted common shares: ShareCur(bars)

You can now access a number of common shares outstanding that may be more frequently updated than line items from financial statements. Please see the documentation for more info. In short this number may be more up-to-date than the last figure from financial statements due to several reasons, including corporate events in the middle of a quarter, buybacks, or the processing of proxy statements.

  1. Two tweaks to the main function that returns number of shares

Our main function to get the number of shares tries very hard not to return N/A. It has several fallback mechanisms when the desired number of shares is unavailabe (for whatever reason, including preliminary reports or major re-orgs). For example, if non-diluted common shares (as used in MktCap) is needed and the value is N/A then it tries to get the value from previous quarter. Failing that it returns the # of shares from the diluted line item.

The tweaks are these:

i) When non-diluted , most current shares is N/A it will return the value from the new function SharesCur(0)

ii) When diluted, most current shares results in a number less than SharesCur(0) , then SharesCur(0) is returned

  1. Short interest as % of shares (SI%ShsOut) now returns N/A if for any reason it evaluates to a percent greater than 99%.

  2. We noticed instances of PE being NA because earnings per share was N/A, however there was a valid value for Income Available to Common (or IAC). In this case we will calculate EPS ourselves and return PE.

These changes were made due to observations of some invalid ratios.

For example, Michaels Companies Inc (MIK) market cap was being calculated as 1.8B rather than 3B because it just went through a stock re-issuance and the number of shares from the financial is N/A. It is falling back to a previous value which was from the no-longer-existing shares. It is now correct since it uses the share figure from SharesCur.

Another example was from a Short Interest % being more than 100%. This was due to an incorrect value from Compustat which has been reported (and should correct itself soon). However in no instance does it make sense to return a value greater than 100%.

We don’t think any of these changes should impact your simulations.

Thank You

This is good! Which share count does MktCap use?

MktCap uses latest non diluted shares line item from interim filings, so there should be no changes in 99.??% of the cases. If N/A it will try to use SharesCur(0), then fall through to other options. See the docs on “Shares” for the steps taken.

Props, Marco. This newer, better shares will help us.

What about Float? Any idea how, and how timely, this is calculated?

Thanks for this update, the more reliable and timely the data the better our models will perform [=D>]