Variance measures the average squared deviation of observations from their mean. In financial analysis, it summarizes how widely prices or returns fluctuate, but its squared units make it harder to interpret directly. section 3.4.2 section 7.2.2
Use variance with these practical distinctions:
- Pair variance with standard deviation: standard deviation is the square root of variance and returns dispersion to the original units. section 3.4.2 section 7.2.2
- Prefer returns rather than absolute share prices when comparing stocks; price levels do not provide a reliable performance comparison. section 7.2.2
- Use variance for theoretical statistics and relative dispersion, while relying more often on standard deviation in practical analysis. section 7.2.2
Computing variance for returns
A pandas workflow aggregates mean, standard deviation, and variance from log returns: `summary = log_rets_2023.agg(["mean", "std", "var"]).T`. For 2023, the reported variance is 0.018214 for Coca-Cola, 0.039690 for Apple, and 0.639385 for NuScale, indicating much greater dispersion for NuScale. section 3.4.2 section 3.4.2
