12
Two-Period Panel Data Analysis According to CANA, more people are choosing cremation because it is (1) affordable, (2) environmentally friendly, (3) easier given our โ€œgeography and population mobility, โ€ and (4) increasingly acceptable to religious groups. http://www.kates-boylston.com/NewsPage.aspx? newsID=2122 Cremation rate (%) Natives, born in state (%) Year dummy The 35 states are: AL, AZ, AR, CO, CT, FL, GA, ID, IN, IA, KS, KY, ME, MD, MA, MI, MN, MO, MT, NE, NV, NJ, NM, NC, ND, OR, PA, SC, SD, TX, UT, VT, WA, WI, and WY = 0 + 0 2000 + 1 + A simple empirical specification that focuses on Boylstonโ€™s third explanation for the increasing proportion of people choosing cremation is: ( 1 ) Pooled OLS

Two-Period Panel Data Analysis

  • Upload
    conway

  • View
    67

  • Download
    0

Embed Size (px)

DESCRIPTION

Two-Period Panel Data Analysis. Pooled OLS. According to CANA, more people are choosing cremation because it is (1) affordable, (2) environmentally friendly, (3) easier given our โ€œgeography and population mobility, โ€ and (4) increasingly acceptable to religious groups. . - PowerPoint PPT Presentation

Citation preview

Page 1: Two-Period Panel Data Analysis

Two-Period Panel Data Analysis

According to CANA, more people are choosing cremation because it is (1) affordable, (2) environmentally friendly, (3) easier given our โ€œgeography and population mobility, โ€ and (4) increasingly acceptable to religious groups.

http://www.kates-boylston.com/NewsPage.aspx?newsID=2122

Cremation rate (%)

Natives, born in state (%)Year dummy

The 35 states are: AL, AZ, AR, CO, CT, FL, GA, ID, IN, IA, KS, KY, ME, MD, MA, MI, MN, MO, MT, NE, NV, NJ, NM, NC, ND, OR, PA, SC, SD, TX, UT, VT, WA, WI, and WY

๐ถ๐‘…๐‘–๐‘ก=๐›ฝ0+๐›ฟ0๐‘Œ๐‘Ÿ 2000+๐›ฝ1๐‘ ๐‘–๐‘ก+๐œ€๐‘–๐‘ก

A simple empirical specification that focuses on Boylstonโ€™s third explanation for the increasing proportion of people choosing cremation is:

(1)

Pooled OLS

Page 2: Two-Period Panel Data Analysis

๐ถ๐‘…๐‘–๐‘ก=64.4+9.34 โˆ™๐‘Œ๐‘Ÿ 2000โˆ’0.757 โˆ™๐‘ ๐‘–๐‘ก

0 10 20 30 40 50 60 70 80 90 100051015202530354045505560657075

Cremation

Rate (%)

Native (% born in state)

๐‘ ๐‘™๐‘œ๐‘๐‘’=โˆ’0.757๐ถ๐‘…

๐‘–๐‘ก=2000๐ถ๐‘…

๐‘–๐‘ก=1990

9.34

Page 3: Two-Period Panel Data Analysis

0 10 20 30 40 50 60 70 80 90 100051015202530354045505560657075

Cremation

Rate (%)

Native (% born in state)

Colorado

Georgia

Fewer people living in Colorado were born there than in Georgia and a lot of the variation in Native used to estimate is coming from between states and some of the variation is coming from within states over time.

Page 4: Two-Period Panel Data Analysis

State fixed effect () captures (time-invariant and unobserved) prices, regulations, environmental attitudes, religious attitudes. If they are observable, you are better off putting them into the equation as explanatory variables.

๐œ€๐‘–๐‘ก=๐‘Ž๐‘–+๐‘ข๐‘–๐‘ก

Time varying error (idiosyncratic error) โ€”unobserved factors that affect cremation rates and vary over time

๐ถ๐‘…๐‘–๐‘ก=๐›ฝ0+๐›ฟ0๐‘Œ๐‘Ÿ 2000+๐›ฝ1๐‘ ๐‘–๐‘ก+๐œ€๐‘–๐‘ก

๐ถ๐‘…๐‘–๐‘ก=๐›ฝ0+๐›ฟ0๐‘Œ๐‘Ÿ 2000+๐›ฝ1๐‘ ๐‘–๐‘ก+๐‘Ž๐‘–+๐‘ข๐‘–๐‘ก

Fixed Effects Model

(2)

(1)

Pooled OLS

Pooled OLS is not substantially different from single-time-period OLS. If you have an omitted variable problem due to stuff in the error term, pooling the data doesnโ€™t eliminate it.

Page 5: Two-Period Panel Data Analysis

๐ถ๐‘…๐‘–๐‘ก=๐›ฝ0+๐›ฟ0๐‘Œ๐‘Ÿ 07+๐›ฝ1 ๐‘ ๐‘–๐‘ก+๐œ€๐‘–๐‘ก

For simplicity, suppose

๐ถ๐‘…๐‘–๐‘ก=๐›ฝ0+๐›ฟ0๐‘Œ๐‘Ÿ 07+๐›ฝ1 ๐‘ ๐‘–๐‘ก+๐›ฝ2 ๐‘…2๐ธ๐‘–๐‘ก+๐‘ข๐‘–๐‘ก

Eโƒ๏ฟฝ โ€“ โƒ๏ฟฝ โ€“ โƒ๏ฟฝ+

downward bias

Page 6: Two-Period Panel Data Analysis

๐ถ๐‘…๐‘–2000=๐›ฝ0+๐›ฟ0+๐›ฝ1 ๐‘ ๐‘–2000+๐‘Ž๐‘–+๐‘ข๐‘– 2000

๐ถ๐‘…๐‘–1990=๐›ฝ0+๐›ฝ1 ๐‘ ๐‘–1990+๐‘Ž๐‘–+๐‘ข๐‘–1990

โˆ†๐ถ๐‘…๐‘–=๐›ฟ0+๐›ฝ1โˆ†๐‘ ๐‘–+โˆ†๐‘ข๐‘–

First-difference equation: eliminates

First Differences

(3)

Page 7: Two-Period Panel Data Analysis

Estimating Fixed Effects Models

๐ถ๐‘…๐‘–๐‘ก=๐›ฝ0+๐›ฟ0๐‘Œ๐‘Ÿ 2000+๐›ฝ1๐‘ ๐‘–๐‘ก+๐‘Ž๐‘–+๐‘ข๐‘–๐‘ก(2)

Page 8: Two-Period Panel Data Analysis

Estimating First-Differencing Models

โˆ†๐ถ๐‘…๐‘–=๐›ฟ0+๐›ฝ1โˆ†๐‘ ๐‘–+โˆ†๐‘ข๐‘–(3)

Page 9: Two-Period Panel Data Analysis

Estimating First-Differencing Models โˆ†๐ถ๐‘…๐‘–=๐›ฟ0+๐›ฝ1โˆ†๐‘ ๐‘–+โˆ†๐‘ข๐‘–(3)

Estimating Fixed Effects Models

๐ถ๐‘…๐‘–๐‘ก=๐›ฝ0+๐›ฟ0๐‘Œ๐‘Ÿ 2000+๐›ฝ1๐‘ ๐‘–๐‘ก+๐‘Ž๐‘–+๐‘ข๐‘–๐‘ก(2)

๏ฟฝฬ‚๏ฟฝ0

๏ฟฝฬ‚๏ฟฝ1

๏ฟฝฬ‚๏ฟฝ0

๏ฟฝฬ‚๏ฟฝ1

Demonstrates that models using fixed effects are using variation within states (or cities, counties, colleges, etc.) to estimate parameters

Page 10: Two-Period Panel Data Analysis

๐ถ๐‘…๐‘–๐‘ก=๐›ฝ0+๐›ฟ0๐‘Œ๐‘Ÿ 07+๐›ฝ1 ๐‘ ๐‘–๐‘ก+๐œ€๐‘–๐‘ก

๐œ€๐‘–๐‘ก=๐‘Ž๐‘–+๐‘ข๐‘–๐‘ก

โˆ†๐ถ๐‘…๐‘–=๐›ฟ0+๐›ฝ1โˆ†๐‘ ๐‘–+โˆ†๐‘ข๐‘–

Key Assumption is uncorrelated with

This assumption holds if the idiosyncratic error (u) at each time period is uncorrelated with the explanatory variable in both time periods.

Page 11: Two-Period Panel Data Analysis
Page 12: Two-Period Panel Data Analysis

Costs and Benefits of Fixed Effects Model

Benefitโ€”controls for unobserved factors that vary across states, cities, collegesโ€ฆ Costs1. More expensive data collection2. Can reduce or eliminate variation in explanatory variables.