Extra Flashcards
Why can investment grade bonds not be used to diversify stocks well?
The risk is not high enough, as such even if they are negatively correlated with stocks the return difference will not be enough to offset the stocks loss.
How many shares do we have to buy to hedge selling an option contract for 100 shares that is in the money.
we need to buy deltathe amount of shares. An at the money option has a delta of 0.5, out of the money is less than 0.5 and in the money has a delta of more than 0.5. So 1000.5 = 50 shares.
If one stock decreases in price by 30% over 3 months and one increases in price by 30% over three months what is their price correlation and return correlation?
Price correlation is significantly negative over that period, the return correlation is impossible to know from the given information.
What errors do the traditional sample VCV and correlation have primarily?
Traditional sample is primarily estimation error due to a lack of structure. The correlation is primarily specification error on account of its large structure component.
what does RT AA = αP − λRω2
P − T C measure?
Return out-performance adjusted for risk and transaction costs.
How do I calculate the standard error of an autocorrelation from a time series?
The denominator of the t series for correlation is the standard error.
Suppose you download two years of daily data on all stocks listed on the NZX. For what proportion of these stocks would you expect that you could reject the null hypothesis that the returns are drawn from a distribution with zero mean?
A very small proportion on account of mean blur(standard deviation being so high relative to mean.
Will the Spearman rank-order correlation coefficient work in situations where there are high ranked variables at both ends and low ranks in the middle?
No, it will not work with u-shaped patterns.
What even is option delta? how can we get it with numeric calculus?
Option delta is the slope of the option cost, we can find it numerically by taking a very small step in stock price.
How do I get the value of a two-tailed f test from two standard deviations?
Square the standard deviations, then divide one by the other.
What was the approximate equity risk premium in the US from 1963-2013
About 4-6%
Given we find with a t statistic that our sample mean is beyond the critical value and the autocorrelation for the t-statistic is also beyond the critical value. We have 100 daily stock returns for a small stock what can we tell about the distribution of the data
We have no evidence to reject normality, and even if we had a larger sample, the statistically significant results would not lead us to reject normality.
What will the correlation be between a 60/40 stocks/bonds portfolio and all stocks portfolio roughly? Assuming all high quality bonds?
Quite high, about 98%
✸ The general functional form of a Z- or t-statistic is φˆ −φH0 /SE(φˆ) where φ is some population parameter with a null hypothesis value φH0 and a sample estimator φˆ. In this general case, the denominator SE( φˆ) represents what?
the estimated standard deviation of the underlying sample
If I invested for 50 years in t bills and 50 years in S&P 500 what would be the approximate ratio difference?
8 times higher for stocks.
