Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 : Greenblatt's Little Book That Beats The Market -- Ignore unavailable to you. Want to Upgrade?

To: Jurgis Bekepuris who wrote (190)	3/2/2013 12:39:01 AM
From: Shane M	Read Replies (1) \| Respond to of 218

I guess if you like custom data it's more difficult to automate. I do have the database pre-calculate certain specific fields that I want if it's non-standard but derivable from existing data - but if I can't do it within the database I don't do it because ongoing manual maintenance just isn't what I can keep up for that many stocks. Anything beyond that gets done in excel based on what is export. But yes - I'd agree that if you like your own data and can't derive your own data from existing fields in standard databases, then automation of some steps is probably not a way to go. Keeping track of that type of data in dynamic time series is difficult doing things manually. Based on the adjustments that you say you do, there's clearly a limitation to how much work you could have the computer do for you. I agree with what you say though - the value in the automation is not the initial screening process - it's the assistance in routine flagging of things that are "interesting" after the initial screens to draw attention for closer look. if interested - fwiw, to give an idea framework of how the update process I use works - I just looked at my Buffettology spreadsheet and in it I export 107 columns of data for each stock from the database. From there I have about 116 columns of further calcs that help with the automated projections. For manual stuff on the interesting companies I copy the row of data for a particular stock into a separate worksheet where I have a more tweakable/changeable template where certain factors in the future projection are manually tweakable - but I don't save tweaks each time. The Greenblatt spreadsheet was developed much later than the Buffettology approach, but it's based on a similar process - exporting 110 columns for each company and then have another 100 columns of calcs and a few macros to manage the final sorting for me once all the weighting/ranking of various measures is done. It's not very complicated, I think it's just mostly weighting a bunch of different factors that go into the ultimate ranking. (Like I mentioned, the Greenblatt approach I use is modified to include more than ROC and yield. I did this after realizing I still wanted to look at the Greenblatt companies based on factors I valued, and I tried to bubble those with desirable factors closer to the top of the list based on adjustments). And I guess as a last thing I use is something I recently built that I find find very useful: I'm visual, so I have a separate spreadsheet with charts of alot of long term trends on specific factors that I like to look at, but it's a separate deal that I use to just pull up things that are "interesting" so I can quickly glance at what's happened up to this point. anyhow, good luck with it - but I guess to circle back to your initial concerns - I agree with your concern that the more custom adjustments you do to find your interesting companies - the less valuable the automation will likely be to you, and it may not save you that much time in that context.