Flashcards in Chapter 1 Deck (17)
Is data made up of any number including fractions and portions like 0.5 grams or 11.7% return
Discrete data is data that must fall into a certain value, like whole numbers. You cannot have 89.3 people in your office
Ordinal numbers are used to provide positioning/ordering, so 1st is better then 2nd, and 2nd is better then 4th. However, 2nd is not two times better then 4th.
Nominal numbers are numbers that carry no weight and instead are just used to assign qualitative value, so a phone number isn't any better then another, having an office #289 is simply just different then office #290.
Cardinal numbers are numbers that provide weighted value, so a stock price. This is because a price can be double the value of another or half, etc.
Time Series Data
Time series data is data that is collected over a period of time.
Cross-sectional data is data on one or more variables collected at a single point. Prices of several stocks all on the same day.
Panel data has the dimensions of both time series and cross-section, example: daily prices of several stocks over 2 years.
Problems solved with cross-sectional data
The relationship between company size and the return to its investors.
The relationship between a nations GDP and the probability it will default on debt
Problems solved with time series data
The effect on a country's exchange rate of an increase in trade deficit
How the value of a company's stock price varies when it announced the value of dividend payments.
Noisy data has a large amount of random variation which can make uncovering relations and underlying behavior difficult. (Financial data is considered noisy)
Clean data has minimal noise and is the least free of errors
Simple Return Formula
(New Price - Old Price) / Old Price
Continuously Compounded Formula
ln(New Price / Old Price)
Bond Price in given year terms
(Bond Value / CPI Value) * CPI Value Given Year
Inflation Rate Formula
(New CPI - Old CPI) / Old CPI