317 Lecture Notes

What is the the most important aspect of Model Specification?

Garbage Data in = Garbage Results Out

Parsimony

a guiding principle which suggests that we should prefer simpler explanations and solutions over more complex ones, all other things being equal.

Quanitative Variables

Countable / measurable
Measurable in numbers
How many, how much or how often
Examples: Profit, Price, GDP, Age

Qualitative Variables

Descriptive
Can not be counted or measured
Why, how, or what happened
Examples: Recession, War, Gender

Why do we want to generally have an intercept?

– True intercept
– Garbage collector
– Reduces bias in slope estimates

Core OLS Assumptions

1. Independent variables are non-stochastic (fixed)

2. Number of observations > number of independent variables

3. Linearity in the betas

4. Zero mean error

5. No perfect multicollinearity

6. Homoskedasticity (constant error variance)

7. No Autocorrelation

8. Normality of error terms

Why do we log?

Because regression likes straight lines, but real life grows in curves. Logs measures percentages not volume growth.

What Makes a High-Quality Forecast?

1. Accurate

2. Timely

3. Clear & Understandable

4. Relevant

Qualitative Forecasting Methods

(Judgment-Based)

• Delphi method

• Surveys / market research

• Expert opinion

Quanitative Forecasting Methods

Data-Based)

• Time series models

• Associative (causal) models

Composite Forecast

• Combines multiple independent forecasts and uses weights to form one final prediction

Assumptions of an Autoregressive Process

• The true data-generating process is unknown

• Data is generated by a random process

• Past values influence current values

• The same process holds in the future as in the past (needed for forecasting)

What is production

Production is about turning inputs into outputs.

Three Stages of Production Function

Stage 1 – 0 to max APP
Stage 2 – max APP to MPP=0 (Economic Stage)
Stage 3 – beyond MPP = 0

Average Physical Product (APP)

hows the relationship between output and the
quantity of input i used - the average output per unit of
input i.

Marginal Physical Product (MPP)

gives the exact rate of change of the TPP function for
an infinitesimal change in input

In finance terms what is does it mean if the beta is less then, equal or greater to zero?

Picture

Covariance

is a measure of how changes in one variable are associated with changes in another variable.

• Positive → move together

• Negative → move opposite

• Zero → independent

In portfolio diversification there are two types of risk, what are they?

• Systematic – common to all assets

• Unsystematic – specific to individual asset

What needs to happen to reject the null hypothesis in a Chi-Squared proportions and Marasculio test?

the t stat has to be more then the critical value and the p value has to be less then the alpha