Scribd
Upload
22 views

Calculating Critical Value

This document discusses how to calculate critical values for the t-distribution using tables of percentage points. It provides examples of calculating two-tailed and one-tailed critical values at the 5% and 1% significance levels for the t-distribution with 72 degrees of freedom. The calculations involve locating the critical values in the table and interpolating linearly when the degrees of freedom do not match the table values. Formulas for linear interpolation are presented along with the resulting critical values.

Uploaded by

Mitsy Simpson
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
22 views

Calculating Critical Value

This document discusses how to calculate critical values for the t-distribution using tables of percentage points. It provides examples of calculating two-tailed and one-tailed critical values at the 5% and 1% significance levels for the t-distribution with 72 degrees of freedom. The calculations involve locating the critical values in the table and interpolating linearly when the degrees of freedom do not match the table values. Formulas for linear interpolation are presented along with the resulting critical values.

Uploaded by

Mitsy Simpson
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

ECON 351*: Critical Values of t-Distributions M.G.

Abbott

Calculating Critical Values of t-Distributions Using Tables of Percentage Points

Refer to: Table of Percentage Points of the t Distribution, p. 3 of this document.

Two-Tailed Critical Values of the t[72]-Distribution

Two-tailed 5 percent critical values of the t[72] distribution: α = 0.05 ⇒ α/2 = 0.025.

• Locate column of table headed 0.025 ← area in one tail of t-distribution


0.05 ← area in two tails of t-distribution

t0.025[72] lies between t0.025[60] = 2.000 and t0.025[120] = 1.980.

t0.025[72] can be approximated using the following linear interpolation formula:


72 − 60 12
t 0.025 [ 72 ] ≅ 2.000 − (2.000 − 1.980) = 2.000 − 0.02 = 2.000 − 0.004 = 1.996
120 − 60 60

• Result:

♦ Upper α/2 = 0.025 critical value of t[72] distribution = t0.025[72] = 1.996;


♦ Lower α/2 = 0.025 critical value of t[72] distribution = − t0.025[72] = − 1.996.

• Stata invttail(df, α/2) function: display invttail(72, 0.025) returns the value 1.994.

Two-tailed 1 percent critical values of the t[72] distribution: α = 0.01 ⇒ α/2 = 0.005.

• Locate column of table headed 0.005 ← area in one tail of t-distribution


0.010 ← area in two tails of t-distribution

t0.005[72] lies between t0.005[60] = 2.660 and t0.005[120] = 2.617.

t0.005[72] can be approximated using the following linear interpolation formula:


72 − 60 12
t 0.005 [72] ≅ 2.660 − (2.660 − 2.617) = 2.660 − 0.043 = 2.660 − 0.0086 = 2.651
120 − 60 60

• Result:

♦ Upper α/2 = 0.005 critical value of t[72] distribution = t0.005[72] = 2.651;


♦ Lower α/2 = 0.005 critical value of t[72] distribution = − t0.005[72] = − 2.651.

• Stata invttail(df, α/2) function: display invttail(72, 0.005) returns the value 2.646.

ECON 351*: Computing Critical Values of t-Distributions Page 1 of 3 pages


351memo2.doc
ECON 351*: Critical Values of t-Distributions M.G. Abbott

One-Tailed Critical Values of the t[72]-Distribution

One-tailed 5 percent critical value of the t[72] distribution: α = 0.05.

• Locate column of table headed 0.05 ← area in one tail of t-distribution


0.10 ← area in two tails of t-distribution

t0.05[72] lies between t0.05[60] = 1.671 and t0.05[120] = 1.658.

t0.05[72] can be approximated using the following linear interpolation formula:


72 − 60 12
t 0.05 [72] ≅ 1671
. − (1671
. − 1658
. ) = 1671
. − 0.013 = 1671
. − 0.0026 = 1668
.
120 − 60 60

• Result:

♦ For a right tail test, α = 0.05 critical value of t[72] distribution = t0.05[72] = 1.668;
♦ For a left tail test, α = 0.05 critical value of t[72] distribution = − t0.05[72] = − 1.668.

• Stata invttail(df, α) function: display invttail(72, 0.05) returns the value 1.666.

One-tailed 1 percent critical value of the t[72] distribution: α = 0.01.

• Locate column of table headed 0.01 ← area in one tail of t-distribution


0.02 ← area in two tails of t-distribution

t0.01[72] lies between t0.01[60] = 2.390 and t0.01[120] = 2.358.

t0.01[72] can be approximated using the following linear interpolation formula:


72 − 60 12
t 0.005 [72] ≅ 2.390 − (2.390 − 2.358) = 2.390 − 0.032 = 2.390 − 0.0064 = 2.384
120 − 60 60

• Result:

♦ For a right tail test, α = 0.01 critical value of t[72] distribution = t0.01[72] = 2.384;
♦ For a left tail test, α = 0.01 critical value of t[72] distribution = − t0.01[72] = − 2.384.

• Stata invttail(df, α) function: display invttail(72, 0.01) returns the value 2.379.

ECON 351*: Computing Critical Values of t-Distributions Page 2 of 3 pages


351memo2.doc
ECON 351*: Critical Values of t-Distributions M.G. Abbott

Table of Percentage Points of the t Distribution

Source: Damodar N. Gujarati, Basic Econometrics, Third Edition. New York: McGraw-Hill,
1995, p. 809.

ECON 351*: Computing Critical Values of t-Distributions Page 3 of 3 pages


351memo2.doc

You might also like