Hey all! Still planning out my first strategy and had some questions come up that I'd like to get your input on.
In situations where you'd want to buy/sell into a low liquidity stock, are there best practices for how to determine how much to buy/sell for a given time period if you're trying to avoid spikes? My thinking is that it would be some function of the stock's liquidity in the preceding time period, but I'm not sure what a reasonable way to calculate that is.
Along with that, how do you go about accounting for slippage when back testing? I'm planning to buy trade-by-trade data for the stocks I'm interested in, so it seems like I'd have the data necessary to account for it reasonably well. I'm planning to create my own backtests. My thought right now is to calculate a worse case scenario for how long it would take for an order to get filled and then see what the price was at that point. That feels kind of naive, though. Maybe it's better to first account for the time it takes for the order to be received by the broker and put on the market, and then step through the historical trades pretending the trades were my own until I've reached the number I intended to buy? But even that seems kind of naive since in reality there are other people trying to buy as well and I won't be given priority. Maybe there's a better way to handle this?
Thanks in advance for your time!
Submitted September 28, 2020 at 05:28AM by EdvardDashD