Charles Griffin & Company Limited, 1946. - 521 pages.
PREFACE TO VOLUME II
This volume falls into five sections. The first, comprising chapters 17 to 20, deals with Estimation. The second, comprising chapters 21, 23, 24 and 26 to 28, covers the Theory of Statistical Tests, including the Analysis of Variance and Multivariate Analysis. The third, consisting of chapter 22, deals with Regression Analysis and completes the account of statistical relationship begun in chapters 13 to 16 of Volume I. In the fourth, chapter 25, I have tried to give an introductory account of the reaction of theoretical considerations on the Design of Statistical Inquiries. Finally, the fifth, comprising chapters 29 and 30, deals with the Analysis of Time-Series.
The literature of statistical theory is now so vast that it seemed worth while devoting considerable space to a bibliography, which is given in Appendix B. Although it is far from complete, I hope that it will serve its purpose in guiding the student to the main sources.
The chief problem in the writing of this volume arose in connection with the logic of statistical inference. Whenever possible I have kept the treatment objective. It is, I consider, unfair in a book of this kind not to present all sides of a case, particularly when there is so much disagreement among the authorities. Some day I hope to show that this disagreement is more apparent than real, and that all the existing theories of inference in probability differ essentially only in matters of taste in the choice of postulates. But this book is not the place for such work, and for the present I am content to state the position and to leave the reader to exercise his own choice.
PREFACE TO VOLUME II
This volume falls into five sections. The first, comprising chapters 17 to 20, deals with Estimation. The second, comprising chapters 21, 23, 24 and 26 to 28, covers the Theory of Statistical Tests, including the Analysis of Variance and Multivariate Analysis. The third, consisting of chapter 22, deals with Regression Analysis and completes the account of statistical relationship begun in chapters 13 to 16 of Volume I. In the fourth, chapter 25, I have tried to give an introductory account of the reaction of theoretical considerations on the Design of Statistical Inquiries. Finally, the fifth, comprising chapters 29 and 30, deals with the Analysis of Time-Series.
The literature of statistical theory is now so vast that it seemed worth while devoting considerable space to a bibliography, which is given in Appendix B. Although it is far from complete, I hope that it will serve its purpose in guiding the student to the main sources.
The chief problem in the writing of this volume arose in connection with the logic of statistical inference. Whenever possible I have kept the treatment objective. It is, I consider, unfair in a book of this kind not to present all sides of a case, particularly when there is so much disagreement among the authorities. Some day I hope to show that this disagreement is more apparent than real, and that all the existing theories of inference in probability differ essentially only in matters of taste in the choice of postulates. But this book is not the place for such work, and for the present I am content to state the position and to leave the reader to exercise his own choice.