Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
22 Cards in this Set
- Front
- Back
|
Alternative hypothesis Ha |
Hypothesis accepted when the null is rejected |
|
|
Backfill bias |
When backfilled data in a database only includes managers with good perf who are more likely to wat to publicize their results. Thus, the database investment results are biased upwards |
|
|
Backfilling |
The act of inserting an investment into a database along with actual trading history for prior periods |
|
|
Backtesting |
The use of historical data to test a strat that is selected after teh data is observed. Backtesting with data dredgign can generate false indications of future returns |
|
|
Bayesian Formula |
Calcs updated probs of event given new info |
|
|
Causality |
When one variables corr. with another variable is b/c its value determines the value of other variable. The rise in the market may cause a long only fund to rise in teh value but the long only fund is probably not causing another long only fund to rise in value. Their respective rises are cause by the rise in the market |
|
|
Cherry picking |
Process of extracting only those reults that support a certain point of view. An example would be a fund manager only publiciing instances when his strat worked and not disclosing when it failed |
|
|
Confidence level |
The probs that a result may Not be due to randomness. Usually represented as a large probability (90%, 95% or 99%) and equals 100% minus the significance level |
|
|
Data dredging |
Refers to the overuse and misuse of stat tests to identify historical pattersn. The prob with data dredging is that it fails to take the number of tests performed into account with analysing the results |
|
|
Economic significance |
The extent to which a variable in an economic model has a meaningful impact on another variable in a practical sense |
|
|
Null hypothesis H0 |
The hypothesis to be tested. The null hypothesis is considered True unless the hypthesis test gives convincing evidence that the null is False |
|
|
Outlier |
Values that are unusally large or small. May influence the results of regression and the est. of the corr. coefficient. Excluding outliers may reduce the corr. number |
|
|
Overfitting |
Small changes in the stat inputs can cause large changes in the min-var and efficient frontier |
|
|
p-Value |
Smalles level of significance at which the null hypothesis can be rejected. The smalled ther p-val the stronger the evidence against the null hypothesis and in favor of the alternative hypothesis |
|
|
Selection bias |
Distortion in the sample selected resulting from the method used to select them |
|
|
Self selection bias |
Results in an upward bias when funds that have performed well are more likely to report their results |
|
|
Significance level |
The probs that a result may be due to randomness. Typically represented as the small probs (1,5,10%) that a result was falsely generated by randomness |
|
|
Spurious correlation |
It means that there may appear to be a relationship between 2 variables when in fact there is NO economic explaination for a relationship. 2 vars may not be correlated but may show spurious corr when a 3rd var is introduced. |
|
|
Survivorship bias |
Upward bias from the deletion of the historical performance of funds ceasisng to report due to liq, failure or closure |
|
|
Test stat |
The variable that is analyzed in order to make the decision whether to reject or fail to reject the null hypotheis |
|
|
Type 1 error |
Falsely rejecting a true null |
|
|
Type II error |
Failing to reject a false null |
