When you buy cheap stocks, you should consider that they may be on their way to becomming free (i.e. non-survivors). If your database does not contain stocks that are no longer traded (because the company merged or went out of business), your test results will be unrealistically optimistic. Lookahead bias occurs any time you use data that was unavailable at the time you would have traded. Quarterly and annual financial reports are never available at the end of the querter/year, and they often get restated later. Restatements never make the company/stock look better than before. You need to know when the data was first available and what it looked like then.
Interesting, thanks to you and also to dcraig. Since you seem knowledgeable on the topic, I'll ask you whether you know if Robert Schiller took such biases into account for this data plot: http://en.wikipedia.org/wiki/File:P...tor_of_Twenty-Year_Returns_(Shiller_Data).png Generally he's highly thought of, so I figure he would have ... right?
I don't really have an opinion, but I tend to be a bit skeptical about stock price data going back to the 19th century. The basic problem I have with long-term investing is that by the time you have usable feedback, you're dead.
That plot could be entitled "Growth is a Myth" Although given the lack of seriously negative results (ie. bankruptcies) it almost has to suffer from some sort of survivorship bias.
there are tons of resources on value investing I think long term investing is more about purchasing a business. Too many think value investing is find this one stock, that is trading at book value, with .50 cents on the dollar in cash. I think those are one in a million, but I see value investing about buying the right business and paying the right price. The father of value investing was Ben Grahm Buffets mentor and pushed the concept of quantitave analysis. But even Buffet has said I would rather pay a fair price for a great business then a great price for a fair business. actually Buffet's partner Munger said it. People buy businesses all the time, and your not looking to get a 50% discount. When you put the power of dividends reinvesment, compunding and a steady growth it is a powerful tool. I think a well rounded investor can move from short term to long term and back. http://www.oldschoolvalue.com/ http://www.moderngraham.com/ http://seekingalpha.com/author/dividends4life http://www.gurufocus.com
Since we're mentioning Ben Graham, it should be noted that his definition of the "intrinsic value" of an equity position is the discounted value of the future dividends (or other disbursements). By that definition, the intrinsic value of most modern common stock is nearly or exactly zero. So he would probably have agreed with most of what I've said. Too many people think "value investing" is buying stock and owning it for a long time.
A computer cannot look at a a list of patents in a companies portfolio and assign a future value. This is where you need to go and have a good understanding. Value investing is not just looking at balance sheets, but at management, patents if any, what the company is doing, where they are going etc.. All this requires yourself to make the evaluation. If after evaluating all this and you feel the franchise is a good investment then you go ahead and put some capital at risk for fractional ownership in the company.
I do not see the point of owning non dividend paying stock. That is like owning a farm and having to sell parts of your farm every year until you run out of farmland to sell. To me dividend paying stock is the equivalent of buying a farm that produces food that is sold and the proceeds could be used to reinvest in more acreage (ie dividend reinvest, more stocks) or to be spend as cash instead.
I agree long term investing in non dividend paying stocks should be a minor concentration. As pure home run plays. trying to grab a microsoft or google. but that is not value investing since these companies in their early stages will appear overvalued. Established companies pay dividends.
P/Es can be almost worthless because they are historical. It is future earnings that matter, not past earnings. Ditto for PEG and, to a lesser extent, dividends. Relying on things like P/E or dividend yields is how a lot of non-value investors lose fortunes getting suckered into value traps, like Bill Miller with Fannie Mae (meanwhile I was shorting it down to low single digits despite being a value investor myself) or Mohnish Pabrai with that financial stock that went broke. A value investment is something that i) has practically no chance of going broke ii) is so cheap that even in the worst case scenario, you will make money. So for example, a company which owns $100 million of real estate, has $40 million of long term debt, and is selling for a market cap of $10 million, with a dividend yield of 30%, would be a value investment. True value investments tend to only present themselves near bear market lows, or when individual companies or sectors are in serious short-term trouble, or when a stock/sector/market/country has been ignored for years.