An Overview of Economic Data Su, Chapter 2, section I.

An Overview of Economic Data

Su, Chapter 2, section I

Classifications of Economic Data

• Many ways to classify economic data

• Most depend on some observable feature

• Classification helps to put the data in context

Primary vs. Secondary Data

• Primary: First Hand. Generated by researcher in experiments. Common in physical sciences. Can be replicated

• Secondary: Collected by another party, typically the government. – Non-experimental– Observational– Needs careful treatment

Time Series vs. Cross Sectional

• Time Series - “Dynamic”– Generated sequentially through time

– A measurement of the same economic agent (individual, firm, etc.)

– Dynamic

• Cross Sectional - “Static”– Generated across different agents

– At same point in time

– Static

Panel Data

• Combination of cross-sections over time

• Combines features of both types

Macroeconomic vs. Microeconomic

• Depends on level of aggregation

• Macro– Measures activity of entire economy– Reflects many agents behavior

• Micro– Measures activity of individuals

High-Frequency vs. Low Frequency

• Time Series and Panel distinction only

• Most series are “discrete”

• Categories– Real Time– High Frequency– Low Frequency

• Hourly, Daily, Weekly, Monthly, Annual

Frequency Considerations

• Little high frequency economic data exist

• Must use care when mixing frequencies

• Low frequency data can be treated as high frequency data aggregated over time

• Information is lost as high frequency data are aggregated

Stocks vs. Flows

• Stock data: Measures data at a point in time– Capital stock, end of month checking balance

• Flow data: Occurs over a period of time– Investment, monthly earnings and spending

Units of Measurement

• Data in levels: Absolute size of these variables convey sense of magnitude– Billions of Dollars

– Millions of persons

• Data in changes: Sense of dynamics. Ratios.– Growth rates

– Differences

• Economic data are typically expressed in dollar terms

Real vs. Nominal

• Data measured in dollars must be corrected for changes in prices

• Ruler of variable length

Indexes

• Relative comparison of values• Commonly used

– Price Indexes

– Dow Jones Index

– Industrial Production

• Laspeyers: t = (PtxQo)/(PoxQo)x100– Base Weighted

• Paasche: t = (PtxQt)/(PoxQt)x100– Current weighted

Ratios

• Growth Rate of • GDP = (GDPt - GDPt-1)/GDPt-1x100

• Inflation = (Pt - Pt-1)/Pt-1x100

• Interest = (Dt - Dt-1)/Dt-1x100 D: Deposits

• Unemployment Rate = (LF-EMP)/LFx100• LF: Labor Force

• EMP: Level of employment

Historical Movements

• Use Plots to depict movements– “Time Shape” or “Time Path”

• Generate characteristics or “summary statistics”

• Conveys much information

Quarterly GDP 1947 - 1967

1000

1500

2000

2500

3000

35001947

1948

1950

1952

1954

1955

1957

1959

1961

1962

1964

1966

30 Year Mortgage Rates 1971-1991

0

5

10

15

20

1971

1972

1974

1975

1977

1978

1980

1981

1983

1984

1986

1987

1989

1990

Monthly Unemployment Rate

024681012

1970

1971

1973

1974

1976

1977

1979

1981

1982

1984

1985

1987

1989

1990

First-Order Difference

• Or “First Difference”

Yt = Yt - Yt-1 for all t

• Change in Y

• If negative, Y fell

• If positive, Y rose

• Like Growth RateGR(Yt) = (Yt - Yt-1)/Yt-1 = Yt/Yt-1

First Difference of Quarterly GDP

-100

-50

0

50

100

150

1947

1948

1950

1951

1953

1954

1956

1957

1959

1960

1962

1963

1965

1966

Growth in Quarterly GDP

-20

-10

0

10

20

301947

1948

1950

1951

1953

1954

1956

1957

1959

1960

1962

1963

1965

1966

First Difference of Mortgage Rate

-3

-2

-1

0

1

2

3

1971

1972

1974

1975

1976

1978

1979

1980

1982

1983

1984

1986

1987

1988

1990

1991

Second-Order Difference

• Or “Second Difference”

Yt = Yt - Yt-1 = Yt - 2Yt-1 + Yt-2

• Change in Y

Second Difference of Quarterly GDP

-75

-25

25

75

125

1947

1949

1950

1952

1953

1955

1956

1958

1959

1961

1962

1964

1965

1967