Wrong Answers Flashcards
Stochastic oscillator
Stochastic oscillator measures the relationship between close, high, and low prices
Lagging sectors
The sectors that are linked with commodity and technology stocks are referee to as “lagging sectors.”
Relative strength index
A subtype of momentum oscillators
Hypothesis Testing Errors
- -Failing to reject H0 when it is actually true is not an error, it is correct
- -Rejecting H0 when it’s true is Type 1
- -Accepting when it’s not true is Type-II error
The smaller the significance level, the smaller the probability of making
A Type 1 error, and the greater the probability of making a type-II error
Chi-square
In tests concerning variance of single, normally distributed population, the test statistic most likely used is chi-square with n-1 degree of freedom where n is sample size
Sample Selection Bias
—Most common form is survivorship bias
—Time-period bias occurs when data gathered is for period that is either too short or too long
—Look-ahead bias occurs when a relationship is tested by using a sample data that was not available on the test date.
Central Limit Theorem
According to the central limit theorem the distribution of sample mean has following properties, as long as sample mean is large.
- -Distribution of sample mean will be approximately normal
- -Distribution mean will be equal to population mean from which samples are drawn
- -Variance of distribution of sample mean will be equal to variance of population mean divided by sample siE
Panel Data
Used for single characteristic,ultisols observation units at different times.
Efficiency estimator.
1) An unbiased estimator with the smallest variance is referred to as an efficient estimator.
2) Efficiency of an estimator is measured by it’s variance.
As degree of confidence increases:
the confidence interval widens & precision in estimating the population parameter lowers.
Z-alternative
- -Large sample and unknown population variance
- -Confidence interval calculated as average +sd/sqrt n
Self-selection bias
occurs when hedge funds with poor track records do not voluntarily disclose
Data mining bias
–Detected by conducting out-of-sample tests of the proposed variable.
–When the variable is not statistically significant in out-of-sample tests, it indicates that the variable suffers from data mining bias.
Stratified random sampling
Used in pure bond indexing or in full replication approach in which an investor wants to fully replicate the index by owning all the bonds in the index in proportion to their market value.
Continuous uniform probability distribution
Defined over a range, which sets the parameters for the distribution
Bernoulli variable
- -A binomial variable, which is random
- -number of trials is 1
Probability less than or equal to
Given value-Lower bound/
upper value -lower bound
Assumptions of binomial distribution
(1) Probability of success is constant for all trials
(2) The trials are independent.
Lognormal distribution
- -Longnormal random variables cannot be negative unlike normal distribution
- -Like normal distribution, lognormal distribution is complete described by two parameters mean and variance
- -Lognormal distribution is bounded below by zero and skews to the right
Roy’s Safefy-first ratio
Expected return-threshold/sd
Money market yield
360 x BDY/360-(t x BDY)
Money market yield from Market prices
BD=(Face value- Market price)/Face value
x
360
/
days to maturity
Money-weighted rate of return
The internal rate of return taking account of all cash flows