#
In re: High-Tech Employee Antitrust Litigation

### Filing
716

Omnibus Declaration of Christina J. Brown in Support of #715 Reply re Joint Motion to Exclude the Expert Testimony of Edward E. Leamer, Ph.D. , #714 Reply to Joint Motion to Strike the Improper Rebuttal Testimony in Dr. Leamer's Reply Expert Report or, in the Alternative, MOTION for Leave to Submit a Reply Report of Dr. Stiroh filed by Apple Inc.. (Attachments: #1 Exhibit A, #2 Exhibit B, #3 Exhibit C, #4 Exhibit D, #5 Exhibit E, #6 Exhibit F, #7 Exhibit G, #8 Exhibit H, #9 Exhibit I, #10 Exhibit J, #11 Exhibit K, #12 Exhibit L, #13 Exhibit M, #14 Exhibit N, #15 Exhibit O, #16 Exhibit P, #17 Exhibit Q)(Related document(s) #715 , #714 ) (Brown, Christina) (Filed on 2/27/2014) Modified text on 2/28/2014 (dhmS, COURT STAFF).

EXHIBIT K
OMNIBUS BROWN DECLARATION
Copyright © 2004
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester,
West Sussex PO19 8SQ, England
Telephone (+44) 1243 779777
Email (for orders and customer service enquiries): cs-books@wiley.co.uk
Visit our Home Page on www.wileyeurope.com or www.wiley.com
All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or
transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning
or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the
terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road,
London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the
Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium,
Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to permreq@wiley.co.uk,
or faxed to (+44) 1243 770620.
This publication is designed to provide accurate and authoritative information in regard to the
subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering
professional services. If professional advice or other expert assistance is required, the services of a
competent professional should be sought.
Other Wiley Editorial Ofﬁces
John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA
Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA
Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany
John Wiley & Sons Australia Ltd, 33 Park Road, Milton, Queensland 4064, Australia
John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809
John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, Canada M9W 1L1
Wiley also publishes its books in a variety of electronic formats. Some content that appears
in print may not be available in electronic books.
Library of Congress Cataloging-in-Publication Data
Verbeek, Marno.
A guide to modern econometrics / Marno Verbeek. – 2nd ed.
p. cm.
Includes bibliographical references and index.
ISBN 0-470-85773-0 (pbk. : alk. paper)
1. Econometrics. 2. Regression analysis. I. Title.
HB139.V465 2004
330 .01 5195 – dc22
2004004222
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
ISBN 0-470-85773-0
Typeset in 10/12pt Times by Laserwords Private Limited, Chennai, India
Printed and bound in Great Britain by TJ International, Padstow, Cornwall
This book is printed on acid-free paper responsibly manufactured from sustainable forestry
in which at least two trees are planted for each one used for paper production.
AN INTRODUCTION TO LINEAR REGRESSION
24
is a speciﬁc value chosen by the researcher. If this hypothesis is true we know that
the statistic
0
b − βk
tk = k
(2.49)
se(bk )
has a t distribution with N − K degrees of freedom. If the null hypothesis is not true,
0
the alternative hypothesis H1 : βk = βk holds. As there are no unknown values in tk ,
it becomes a test statistic that can be computed from the estimate bk and its standard
error se(bk ). The usual testing strategy is to reject the null hypothesis if tk realizes
a value that is very unlikely if the null hypothesis is true. In this case this means
very large absolute values for tk . To be precise, one rejects the null hypothesis if the
probability of observing a value of |tk | or larger is smaller than a given signiﬁcance
level α, often 5%. From this, one can deﬁne the critical values tN−K;α/2 using
P {|tk | > tN−K;α/2 } = α.
For N − K not too small, these critical values are only slightly larger than those of
the standard normal distribution, for which the two-tailed critical value for α = 0.05
is 1.96. Consequently, at the 5% level the null hypothesis will be rejected if
|tk | > 1.96.
The above test is referred to as a two-sided test because the alternative hypothesis
0
allows for values of βk on both sides of βk . Occasionally, the alternative hypothesis is
one-sided, for example: the expected wage for a man is larger than that for a woman.
0
0
Formally, we deﬁne the null hypothesis as H0 : βk ≤ βk with alternative H1 : βk > βk .
Next we consider the distribution of the test statistic tk at the boundary of the null
0
hypothesis (i.e. under βk = βk , as before) and we reject the null hypothesis if tk is too
large (note that large values for bk lead to large values for tk ). Large negative values
for tk are compatible with the null hypothesis and do not lead to its rejection. Thus for
this one-sided test, the critical value is determined from
P {tk > tN−K;α } = α.
Using the standard normal approximation again, we reject the null hypothesis at the
5% level if
tk > 1.64.
Regression packages typically report the following t-value,
tk =
bk
,
se(bk )
sometimes referred to as the t-ratio, which is the point estimate divided by its standard
error. The t-ratio is the t-statistic one would compute to test the null hypothesis that
βk = 0, which may be a hypothesis that is of economic interest as well. If it is rejected,
it is said that ‘bk differs signiﬁcantly from zero’, or that the corresponding variable
HYPOTHESIS TESTING
31
2.5.7 Size, Power and p-Values
When an hypothesis is statistically tested two types of errors can be made. The ﬁrst
one is that we reject the null hypothesis while it is actually true, and is referred to as a
type I error. The second one, a type II error, is that the null hypothesis is not rejected
while the alternative is true. The probability of a type I error is directly controlled by
the researcher through his choice of the signiﬁcance level α. When a test is performed
at the 5% level, the probability of rejecting the null hypothesis while it is true is 5%.
This probability (signiﬁcance level) is often referred to as the size of the test. The
probability of a type II error depends upon the true parameter values. Intuitively, if the
truth deviates much from the stated null hypothesis, the probability of such an error
will be relatively small, while it will be quite large if the null hypothesis is close to the
truth. The reverse probability, that is, the probability of rejecting the null hypothesis
when it is false, is known as the power of the test. It indicates how ‘powerful’ a test
is in ﬁnding deviations from the null hypothesis (depending upon the true parameter
value). In general, reducing the size of a test will decrease its power, so that there is
a trade-off between type I and type II errors.
Suppose that we are testing the hypothesis that β2 = 0, while its true value is in fact
0.1. It is clear that the probability that we reject the null hypothesis depends upon the
standard error of our OLS estimator b2 and thus, among other things, upon the sample
size. The larger the sample the smaller the standard error and the more likely we are
to reject. This implies that type II errors become increasingly unlikely if we have large
samples. To compensate for this, researchers typically reduce the probability of type
I errors (that is of incorrectly rejecting the null hypothesis) by lowering the size α of
their tests. This explains why in large samples it is more appropriate to choose a size
of 1% or less rather than the ‘traditional’ 5%. Similarly, in very small samples we may
prefer to work with a signiﬁcance level of 10%.
Commonly, the null hypothesis that is chosen is assumed to be true unless there
is convincing evidence of the contrary. This suggests that if a test does not reject,
for whatever reason, we stick to the null hypothesis. This view is not completely
appropriate. A range of alternative hypotheses could be tested (for example β2 = 0,
β2 = 0.1 and β2 = 0.5), with the result that none of them is rejected. Obviously,
concluding that these three null hypotheses are simultaneously true would be ridiculous.
The only appropriate conclusion is that we cannot reject that β2 is 0, nor that it is
0.1 or 0.5. Sometimes, econometric tests are simply not very powerful and very large
sample sizes are needed to reject a given hypothesis.
A ﬁnal probability that plays a role in statistical tests is usually referred to as the
p-value. This p or probability value denotes the minimum size for which the null
hypothesis would still be rejected. It is deﬁned as the probability, under the null, to
ﬁnd a test statistic that (in absolute value) exceeds the value of the statistic that is
computed from the sample. If the p-value is smaller than the signiﬁcance level α, the
null hypothesis is rejected. Many modern software packages supply such p-values and
in this way allow researchers to draw their conclusions without consulting or computing
the appropriate critical values. It also shows the sensitivity of the decision to reject
the null hypothesis, with respect to the choice of signiﬁcance level. For example, a
p-value of 0.08 indicates that the null hypothesis is rejected at the 10% signiﬁcance
level, but not at the 5% level.
**Disclaimer:** Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.

**Why Is My Information Online?**