Yep! This is a question that Tom never answered. When I was first on the list he suggested I do a particular position by hand calculation - don't recall which - and to be sure to use the adjusted close price but this never made any sense given that some positions having no dividends had no adjusted close price and others with good dividends have adjusted prices that exceed inflation rates.
Besides, the adjusted close is not the price at the end of day's trading so what is the real relationship between the adjusted and the actual price back a year or more when you are AIMing a stock? The little testing I have done on this is that seems to accelerate sales as the price rises faster than the real price. Plus, there is the problem that the day you buy, you buy at the day's price, not the adjusted price so if you then use the adjusted which can be massively out of sync. Say you buy at $10/share but the adjusted for that day is $7.50 and the next day $7.55 it would likely trigger a buy signal at a price that has no relationship to the actual price that day or the next. So you put a buy order at the price AIM tells you and it is likely to sit there throughout a bull run.
If you want to get the actual close price from Yahoo, download it to a spreadsheet and there is a column with the actual closing price you can use.
Back when I worked at Yahoo documenting the changes they were making there was a problem, all the prior programmers had been terminated but adequate documentation did not exist for the programming language used, Python. Yahoo, rather than hiring back some of them decided to outsource the programming to a company in India but nobody there knew Python, only PHP, an older language that was sort of the precursor to Python but not all that close.
What a mess, so I'm not surprised that there are some weirdness happening there now, like not showing the daily closing and having math errors like I pointed out in a prior post.
News 
Market Data 
Discover 
