Chapter 7 Flashcards
Primary data used for?
Specific projects in mind
What does ons produce
Financial, economic and social data nationally and regionally
Who produces detailed data on banking flows?
Bank for international settlement
Cross section data?
Data collected at one point in time for multiple subjects
Time series data
Data collected for a single variable over time
Population?
All member in a well defined set
What is a sample
A subset of a population
Parameter
Number that describes the population
What is a statistic
Number that describes the sample
What is an unbiased sample and what makes it unbiased
Sample is representative of population. Sample is random and sufficiently large
What are the 3 probability sampling methods
Random
Systematic
Stratified
What are the 4 non probability sampling methods
Convenience sampling
Judgement sampling
Quota sampling
Snowball sampling
Random sampling
Each member in population has equal likelihood of being chosen
Systematic
Every person in population is given a number and the nth person of population is chosen. Eg every 10th person. Fine as long as no hidden list in population numbering
stratified sampling?
subset of population are created depending on certain characteristics. This ensure also groups are represented in sample
Convenience sampling
Take people closets to you
Judgement sampling
Judge what people will make a representative sample and choose them
Quota sampling
Same as stratified but once subgroups created then convineice or judgement sampling is used
Snowball sampling
Desired characteristics are hard to find so you use initial subjects contacts to find rest of sample.
What problem can arrive from interval end frequency data
Wide intervals can lose data
What is a limitation of continuous data
The accuracy of the measuring equipment
Histograms
Continuous data that is split into intervals where the area of the bar represents the frequency
Scatter diagrams and how are they usually plotted - i e what variable goes on which axis?
Good for relationship between two variables, usually put variable that causes change on x axis and variable that is affected on the y
What is the common logarithmic of 100 and 1000
2 and 3
What does logged data allow us to do?
Plot non linear data as linear
What are the two types of summary data
Central tendency and dispersion data
What number is used for the median value of 300 numbers
The average of 150th and 151st numbers
What is the general rule between arithmetic me and and geometric mean and what affects this
Arithmetic mean is greater than geometric mean and the more volatile the series, the greater the difference between the two.
What causes a positive skew graph
Mean is greater than median which is greater than mode
What is a negative skew graph caused by
Mode is greater then median which is greater than mean
When do we get symetrically distributed graph
Mean = mode = median
What is a relative frequency table
Probability distribution table
What is the spread of data above the mean for perfect symmetrical data
50% of data is larger than the mean.
What factors affect the shape of a bell curve
Mean, describes location of the peak
Standard deviation describes the height and width of curve
What percentages of data lies within 1, 2 and 3 standard deviations of the mean?
68% in 1 sd, 95% in 2 sd and 99% in 3sd
What is total area under a bell curve
100%
Problem with bell curves?
Frequency of extreme events is much higher than what a normal distribution predicts as these events likelihoods have fat tails and normal distribution have thin tails
What does increasing the sample size do for sample mean
Decreases the standard deviation for the sample mean
2 measures of correlation
Rank correlation and simple correlation
Example of correlation null and alternative hypothesis s
Null: there is no correlation between variable
Alternative: there is correlation between variables
What method is used to minimise the sum of all errors
Ordinary least squares.as line of best fit cannot go through all points
What is a characteristic of a line that minimises error
Goes through the x and y mean
What does r squared mean? example of r squared of 34%
Proration of variance in the dependant variable that can be explained by the independent variable. Example= 34% f variability has been accounted for =, 66% of variability is unexplained
What is the false assumption used for extrapolation
Relationship remains linear outside of the sample range and consumer behaviour is unchanged
OLS estimator ideal qualities?
BLUE = best linear unbiased estimator
What does best mean in blue
Least variance of estimator that is linear and unbiased
What does unbiased mean in blue
Mean of sample is equal to mean of population
Characteristics of blue
Expected errors is equal to zero, errors are uncorrelated and errors have equal variances
What creates a bias estimator
Errors correlated
What happens is errors of estimator are not constant
We get heteroskedasctitcy. Still remains unbiased but SD is wrong
What could be a cause of error correlation for estimates
If we have omitted variables
What is irr
Interest rate that makes NPV equal to zero
Investment decision making rules for irr and NPV
IRR greater than cost of cap
Npv greater than zero
Then project goes ahead
What does NPV do that IRR doesn’t
Maximises wealth of investors
What does irr maximise
Investment return
What does irr assume
Cash flows reinvested at same rate as as original project
Problem with irr
Cannot handle multiple discount rates and multiple cash flow sign changes
If project y has higher NPV and lower irr compared to x then…
Y has later cash flows because not able to reinvest money
What type of index is the FTSE 30
Equally weighted geometric index
What index’s are market weighted arithmetic index
FTSE all world indices
Cac 40
Dax
FTSE 100
What type of index do the following use: FTSE all world indices
Cac 40
Dax
FTSE 100
Market weighted arithmetic index
What type of index does the Dow jones use
Dimple aggregation of unweighted share prices for 30 large us blue chip stocks
What type of index does s and p 500 use
Market weighted average
What is the coefficient of variation =
Ratio of SD to the mean of pop
What type of model is market model and what does it model
Bivariate model
Relate return on a security to a constant and return on market
