Scribd Logo
Upload

Welcome to Scribd!

  • 40 views

    Week 11 Lecture

    Uploaded by

    Olana Sufiyan
    This document provides an overview of statistical inference methods for estimating population parameters and hypothesis testing about population means. It discusses point and interval estimation, how to choose sample sizes, calculating confidence intervals, and the steps for hypothesis testing which include stating null and alternative hypotheses, calculating test statistics, formulating decision rules, making decisions, and interpreting results. Examples of hypothesis tests on population means are provided and steps are worked through. The document also introduces the concept of p-values in hypothesis testing and when the null hypothesis would be rejected based on p-values.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Week 11 Lecture

    Uploaded by

    Olana Sufiyan
    40 views17 pages
    This document provides an overview of statistical inference methods for estimating population parameters and hypothesis testing about population means. It discusses point and interval estimation, how to choose sample sizes, calculating confidence intervals, and the steps for hypothesis testing which include stating null and alternative hypotheses, calculating test statistics, formulating decision rules, making decisions, and interpreting results. Examples of hypothesis tests on population means are provided and steps are worked through. The document also introduces the concept of p-values in hypothesis testing and when the null hypothesis would be rejected based on p-values.

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides an overview of statistical inference methods for estimating population parameters and hypothesis testing about population means. It discusses point and interval estimation, how to choose sample sizes, calculating confidence intervals, and the steps for hypothesis testing which include stating null and alternative hypotheses, calculating test statistics, formulating decision rules, making decisions, and interpreting results. Examples of hypothesis tests on population means are provided and steps are worked through. The document also introduces the concept of p-values in hypothesis testing and when the null hypothesis would be rejected based on p-values.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    40 views17 pages

    Week 11 Lecture

    Uploaded by

    Olana Sufiyan
    This document provides an overview of statistical inference methods for estimating population parameters and hypothesis testing about population means. It discusses point and interval estimation, how to choose sample sizes, calculating confidence intervals, and the steps for hypothesis testing which include stating null and alternative hypotheses, calculating test statistics, formulating decision rules, making decisions, and interpreting results. Examples of hypothesis tests on population means are provided and steps are worked through. The document also introduces the concept of p-values in hypothesis testing and when the null hypothesis would be rejected based on p-values.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 17

    WEEK 11

    Inferences about a population mean

    Lecturer: Sanjar Mardonov


    Lecture Outline

    1. Point and interval estimation


    2. Choosing the right sample size
    3. Finding a confidence interval for µ.
    4. Hypothesis testing for µ.
    5. p-value in hypothesis testing
    Statistical inference
    Statistical Inference
    Methods for making inferences about population mean, μ:
    1. Estimation of parameter (point or interval estimate)
    Examples:
    • “What is the average age in Uzbekistan”?
    • “How much time do millennials spend watching TV news a day?”
    • “What is the average cost to spend a month in Hawaii?”

    2. Hypothesis testing
    Examples:
    • “Is the average age in Uzbekistan more than 30 years?”
    • “Do millennials watch TV news less than 15 minutes a day?”
    • “Is average cost to spend a month in Hawaii is more than $6000?”
    Estimation
    Estimationof parameters
    of parameters
    A single number calculated from the sample that estimates a target population parameter is called a point
    estimator.
    Interval estimator - a range of numbers that contain the target parameter with a high degree of
    confidence.
    Interval estimate
    Point estimate
    I am 95% confident
    that  is between
    Mean, ,20
    is & Mean 40
    unknown
    X = 30

    SAMPLE
    Confidence
    ConfidenceInterval
    Interval

    Suppose we have a population with unknown .

    • 100(1-)% Confidence Interval for , 


    known. 
    𝑥̅  z/2*
    𝑛
    Here,  - significance level, e.g. 0.05, 0.10, etc.
    Margin of Error (E or ME)

    • 100(1-)% Confidence Interval for ,  unknown.


    If the sample size is large (at least 30), we can replace  with s (sample
    standard deviation)
    s
    𝑥̅  z/2*
    𝑛
    Estimation
    Exercise of parameters
    8.1
    Laptop use in middle school

    For a sample of 106 students, the researchers reported the following statistics on how many
    minutes per day each student used a laptop for taking notes: x* = 13.2, s = 19.5. Now the
    researchers want to estimate the average amount of time per day laptops are used for taking
    notes for all middle school students across the country.

    Calculate a 95% confidence interval for the


    average usage of laptop in middle schools.
    Exercise
    Solution8.1 to Exercise 8.1
    Since sample size is large (n = 106), and population standard deviation is
    unknown, we substitute  with s.  = 0.05 for 95% Confidence Interval
    s
    𝑥̅  z/2* z/2 = 1.96
    𝑛

    13.2  1.96* 19.5


    +06
    ( 9.49; 16.91) minutes
    Interpretation: We are 95%
    confident that average usage of
    laptops in middle school lies 0.95 /2 = 0.025
    /2 = 0.025
    between 9.49 and 16.91
    minutes.
    Solution
    SampletoSize
    Exercise 8.1
    to Estimate a Population Mean

    Here,
    z – z/2 , obtained from z table
    E – margin of error
    n must be rounded up in
    order not to exceed the
    specified margin of error.
    Sample Size to Estimate a Population Mean
    Solution
    The maximum allowable margin of error, E (or ME), is $100. The value of z/2 for a
    95% level of confidence is 1.96, and the standard deviation is $1000.
    Substituting these in the formula:
    n = (+./6∗+000
    +00
    )2= (19.6)2 = 384.16
    After rounding it up, we get 385, that is the required sample size to meet the
    specifications.
    Solution
    Hypothesis Testing
    A hypothesis is an assumption about the population parameter.
    The Null Hypothesis, H0 – general statement or default position that there is nothing new
    happening.
    Refers to the Status Quo.
    Always contains the “=” sign.

    The Alternative Hypothesis,


    Ha
    Opposite of the null hypothesis.
    Challenges the Status Quo.
    Never contains the sign “=”.
    Hypothesis
    Steps ofTesting
    hypothesis testing
    Step 1. State null and alternative hypotheses:
    Case 1. Two-sided Case 2. Left-sided Case 3. Right-sided
    H 0:  = 0 H 0:   0 H 0:   0
    H a :   0 H a :  < 0 H a :  > 0

    Step 2. Calculate the test statistic, denoted as T.S. (in this case zstat):

    z stat = 𝑥̅–/ 𝑛0 ( this is obtained from sample data)

    Step 3. Formulate the decision rule:


    Reject H0 if:
    Case 1. |zstat|  z/2 Case 2. zstat  -z Case 3. zstat  z

    Find critical z (z/2 or z) from the standard normal table (z table).
    Steps
    Stepsof hypothesis testing
    of hypothesis testing
    Step 4.
    Make a decision.
    Either reject Ho or fail to (do not) reject Ho based on comparison on step 3.
    Interpret the result.
    Provide your final conclusion in the context of the research question.
    Exercise
    Exercise8.2 8.2
    Suppose you are looking for a two-bedroom apartment for rent. Local website www.olx.uz claims average
    cost of rent for two-bedroom apartments is $400 in Tashkent. But you suspect the real number is higher than
    $400. So you randomly selected 36 two-bedroom apartments from that website and calculated sample mean
    to be $415 and sample standard deviation to be $55. Use  = 0.10.
    Conduct a hypothesis test to see whether rent rate is higher than $400.
    Solution.
    Step 1. State null and alternative hypotheses:
    H0:   400
    H a:  >
    400
    =
    𝑥̅ – 0
    =
    415 − 400 = 1.636
    Step 2.
    T.S. z stat
    / 𝑛 55/ : 6
    Calculate the
    test statistic.
    Solution
    Solutionto Exercise 8.2
    to Exercise 8.2
    Step 3. Formulate the decision rule.
    We reject Ho if zstat  z
    z = z0.10 = 1.28
    Step 4. Make a
    decision and interpret
    the result.
    Since zstat = 1.636 > z = 1.28, we reject
    H0. There is a sufficient evidence to claim
    average rent in Tashkent is greater than $400.
    p-value
    p-value

    Decision rule: Reject Ho if p-value  


    This is a universal rule for many statistical tests.

    Refer to the previous example.


    zstat = 1.636
    p-value = P(z > 1.636) = P(z < -
    1.636) = 0.0505 <  = 0.10
    So we reject the null hypothesis.
    The conclusion remains the same.
    Reference
    Reference
    1. Lind et al (ISBN 978-1-260-18750-2), Chapter 9&10.
    2. McClave & Sincich (ISBN 978-0-321-75593-3), Chapter 7 & 8.
    3. Ott & Longnecker (ISBN 978-0-495-01758-5), Chapter 5.
    THE END

    Thank you for your attention!

    You might also like