Author: rgearyiii

Subject: Re: Recent BCC Signals

Date: 7/29/2019

Recommendations: 14

mungofitch: Their eligible list includes many LLCs, LPs, funds, income trusts, ADRs, preferreds, notes, debentures and mortgage bonds. I didn`t even know there was such a thing as an exchange-listed mortgage bond. Neither did I. If anyone sees their mortgage appear among GTR1 screener results, I apologize in advance, and ask that you let me know so that I can make sure such things are excluded from the GTR1 universe! I track all ~24,000 symbols traded on US exchanges (including OTC Pink and Grey) and have an elaborate piece of code that determines security type and primary exchange listing from multiple sources so that I can filter the list down to what`s described at the top of the backtester form: Use root universe of all securities with primary listings on NYSE/AMEX/NASDAQ/Arca/BATS/IEX (excluding preferred stocks, ADRs/ADSs representing preferred stocks, structured products, ETNs, ETDs, warrants, rights, and units bundling excluded securities). I would think that exchange-listed mortgage bonds would be excluded with exchange-traded debts (ETDs),but I haven`t verified this. As of the close of 7/26/2019, there were 3,470 symbols with primary NASDAQ listing. Among those, 3,089 passed the GTR1 universe inclusion criteria. But as you can see from the Bear Catcher II screen`s steps, after selecting only NASDAQ stocks (after 19731217), the screen further eliminates: o Stocks not of types 10, 11, 18 and 48 (step1). o Stocks with less than 252 market days of non-OTC pricing (step2). o Non-primary share classes (step3). As of 7/26/2019, this leaves 2,051 stocks. That`s a massive difference from the WSJ, if they`re using absolutely everything traded on NASDAQ (3,470 symbols). I.e., about 40% of what the WSJ tracks I`d call "junk" that has little to do with measuring the health of the US equities market, which is what Bear Catcher II is for. Step1 eliminates everything Jim noted, though REITs are allowed. (A REIT, by the way, is a type of closed-end fund complying with legislative requirements for certain tax advantages. Typically only real estate investment funds are able to qualify, but in theory, any company can try.) What`s left is just ordinary common stock in US-incorporated operating companies, plus all REITs (whatever their share type, including shares of beneficial interest). This is almost identical to Russell index eligibility (Russell-FTSE have a more sophisticated means of distinguishing domestic and foreign companies based on a range of factors including geographical source of revenue, location of headquarters, place of incorporation, etc). Consequently, I`ve referred to my version of Bear Catcher II as "the Russell-eligible NASDAQ NH-NL" indicator. mungofitch: We know [WSJ, Pinnacle] use intraday prices not closing, because you can watch the highs and lows climb intraday. My Bear Catcher II uses strict new 252-day highs and lows by closing g-price. See http://boards.fool.com/Message.asp?mid=31878464 and http://boards.fool.com/Message.asp?mid=32177298 . Naturally, I made the choices I did based on backtesting, which revealed fully-adjusted (including dividends) and closing prices to work better. There is also the issue of merger survivors. If stocks A and B merged into stock C within the last year, whose price history do you use before the merger, A`s or B`s? In other words, was A or B the survivor? No two sources will ever completely agree on identifying merger survivors. I have my own algorithms for determining merger/de-merger survivors, but GTR1 linearization is designed to make it not matter quite so much, or at least, it hedges my bets: Prior to merger, daily returns essentially track the sum of the market capitalizations of the two (or more) parties to the merger (with cash dividends). If one stock was so much larger than the other before the merger that it was obviously the acquirer, then its daily returns will dominate in the g-price calculation prior to the merger. I doubt that any other data source on the planet does this in their new high/new low calculations. There are currently about 50 NASDAQ stocks that survived mergers in the last 252 market days (though that seems a bit low, so I might not be counting it correctly). Robbie Geary P.S. If anyone does opt to "dig deeper", be sure to ensure that the screening date (shown in red after a GTR1 Screener run) matches the WSJ as-of date when you make your comparisons. Since the GTR1 database isn`t updated until quite late at night and the WSJ data is updated intra-day, you may need to save the WSJ data the night before and run the GTR1 screener the next morning.