Hello. I am having some success using 'broker ratings' as a backtest parameter. But I am wondering if this uses only current ratings as they are now--not as they were in the past? If so this creates a huge hindsight bias..
I.e., is the backtester using a "current rating" parameter AT THE TIME OF EACH PAST PERIOD--or is it merely using the "current rating" of the present day...??
To check this out, I ran a backtest with 1-week periods and with only ONE parameter: '>1 broker rankings for strong sell' for the last 2 months and also for the first 2 months, two years ago.
Backtest over last 2 months (right now) avg stocks: 173
Backtest over first two months (2 yrs ago) avg stocks: 280
I am hopeful that the backtester SEEMS TO BE using ratings CORRECTLY AS THEY WERE THEN. Because if the backtester were only using the latest ratings--this would fix the 'avg stocks' that would pass the screen--I think. Presumably, there might be fewer of these stocks in the past, because some of these stocks might not have existed. But there could not be more--I think. So, I think the backtester correctly seems to be using "current ratings" meaning "as they were then-current in each past period..."
Am I correct in this assumption? Can the 'current broker ratings' safely be used as a backtest parameter? Thank you anyone.