Timeline for Regression on small values
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 10, 2020 at 1:59 | vote | accept | jitmanchan | ||
| Aug 10, 2020 at 1:59 | comment | added | jitmanchan | Ah, got it. Apologies for not getting back earlier - I have been thinking about what you have said and this makes sense to me now. Thanks a ton! | |
| Jul 30, 2020 at 21:43 | comment | added | eSurfsnake | I'm not sure I follow, but think of it this way. Suppose prices keep going up in general The easiest case to think about is a .01% constant daily increase. Now, this very simple return process is easily predictable - it's +.01% daily. But the price itself - which is pretty much a straight lining going up - has a lot of volatility, since the price keeps changing from day-to-day. The best estimate for tomorrow's price is today's price. That's why returns are looked at. | |
| Jul 30, 2020 at 20:37 | comment | added | jitmanchan | Ah, yes. I agree on the part about the false sense of security. I am using some lagged variables as well in my predictive model for the returns (not the stock price) and based on the coefficients, I do get how that could be misleading. That being said, I want to know as to why the same model, with the same indep variables, and the same degree of correlation on both stock price and stock returns offers such a different trend for each of them. Is there a way by which I could try to at least predict the trends well enough (not the magnitude of the returns, just the trend) | |
| Jul 30, 2020 at 20:12 | history | answered | eSurfsnake | CC BY-SA 4.0 |