28741 - RA Fisher and Statistics Paul R. Earl...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
RA Fisher and Statistics Paul R. Earl pearl@dsi.uanl.mx Facultad de Ciencias Biológicas Universidad Autónoma de Nuevo León San Nicolás, NL, 66450, Mexico
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
A view of the famous biometrician Sir Ronald Aylmer Fisher (1890-1962) is the architect of multivariate analysis. BIOMETRICS of June 1964, Vol 20, No 2 is dedicated to him. Several books are now strongly recommended: An Introduction to Probability: Theory and its Applicaions by William Feller, Time Series Analysis, Forecasting and Control by George EP Box and Gwilym M Jenkins. See JP Benzecri (1982) Bordas, Paris. This much information is enough for a start. Be certain to read R. A. Fisher : The Life of a Scientist. Wiley, New York by Joan Fisher Box, 1978. Another enjoyable book is Choice and Chance by WA Whitworth, 1901 reprinted in 1942 by GE Stechert, New York.
Background image of page 2
Regardless, statistics poses a severe problem for many Latin students. The fault can be poor or little teaching of arithmetic and algebra. Students that fear mathematics cannot succeed in probability, statistics and so forth. On the other hand, students in engineering likely do very well with internet and math. Those students want to learn ! Regardless, the original public school problem seems to be slow reading. The first problem is data management. The students may not know how to enter variables X1, X2, X3. .. They might use 10 columns (cols) for sex which is 1 male, 2 female using up ONLY one col. Suppose X5 in letters takes 10-25 cols. Assign a number. Then this variable will need only 1-3 cols. This seems all very simple--but it's not !
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Biostatistics lecture references See http://www.pitt.edu/~super1/lecture/lec25191/001.htm on analysis of variance. See http://statgen.iop.kcl.ac.uk/bgim/mle/sslike_1.html on maximum likelihood,
Background image of page 4
Analysis of variance (Fisher's ANOVA) Three conceptual models of ANOVA are 1) Fixed-effects model, 2) Random-effects model and 3) Mixed effects, depending most on the number of treatments and levels of the experiment. Oneway gives differences among independent sets. Factorial can help to explain the effects of 2 or more treatment variables. often using 2 X 2 design. Multivariate analysis (MANOVA) is used with more than one dependent variables.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The total sum of squares is partioned into components related into the effects of the model. The number of degrees of freedom (df) can also be partioned and specifies the chi-squared distribution which describes the associated sum of squares. Source
Background image of page 6
Image of page 7
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 21

28741 - RA Fisher and Statistics Paul R. Earl...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online