550.413: Applied Statistics and Data Analysis
Lecture 7: Diagnostics and Remedial Measures II
Applied Mathematics and Statistic, Johns Hopkins University
Finding unusual observations
1
Some observations do not fit the model well. These are
outliers.
2
Som
550.413: Applied Statistics and Data Analysis
Lecture 1: Introduction to Simple Regression
Department of Applied Mathematics and Statistics
Johns Hopkins University
Fall 2016
Course syllabus and administrative details
Instructor: Minh Tang ([email protected]
550.413
Week 1: R basics
Ting Chao
[email protected]
9/9/2016
I. Installation of R
II. RStudio
The interface of RStudio is more friendly but the grammar has no difference
with the original R. The codes below are all run in the RStudio.
(It can be downloaded
550.413: Applied Statistics and Data Analysis
Lecture 6: Linear Models
Applied Mathematics and Statistic, Johns Hopkins University
Box on models
Statisticians, like artists, have the bad habit of falling
in love with their models.
G. E. P. Box
Linear mod
550.413: Applied Statistics and Data Analysis
Lecture 3: Diagnostics and Remedial Measures
Applied Mathematics and Statistic, Johns Hopkins University
Gauss-Markov conditions
Theorem
Under the regression model Yi = 0 + 1 Xi + i where E[i ] = 0,
Var[i ] =
550.413: Applied Statistics and Data Analysis
Lecture 4: Simultaneous inferences and other topics
Applied Mathematics and Statistic, Johns Hopkins University
Bonferroni correction
The Bonferroni correction is a general procedure for simultaneous
inference
550.413: Applied Statistics and Data Analysis
Lecture 2: Inferences in Regression and Correlation Analysis
Department of Applied Mathematics and Statistic
the Johns Hopkins University
Fall 2016
Normally distributed random variables
A random variable X is
550.413 Midterm I Solution
October 15, 2014
Instruction: Please write your name on the first page of the problem set before
you begin. This exam consists of 5 problems. The problems are worth a combined total of 75 points. An additional 15 points (problem
Modern regression 1: Ridge regression
Ryan Tibshirani
Data Mining: 36-462/36-662
March 19 2013
Optional reading: ISL 6.2.1, ESL 3.4.1
1
Reminder: shortcomings of linear regression
Last time we talked about:
1. Predictive ability: recall that we can decomp
550.413 Midterm I
October 10, 2016
Instruction: Please write your name on the first page of the problem set
before you begin. This exam consists of 5 problems. The problems are worth
a combined total of 90 points. You have a total of 75 minutes to complet
Chapter
The
2
Basic
Theory
of
Interest
1. (A nice inheritance) Use the "72 rule". Years = 1994-1776 = 218 years.
(a) i = 3.3%. Years required for inheritance to double = Zf = 8 :'=!21.8. Times
doubled= Hi = 10 times. $1 invested in 1776 is worth 210 :'=!$
550.420
Introduction to Probability
Fall 2014
Midterm Examination # 2
Monday November 3, 2014
=
NAME (Please print clearly):
SECTION:
I agree to complete this examination without unauthorized assistance from any person, materials, or device.
Signature:
=
550.420
Introduction to Probability
Fall 2014
Final Examination
9 AM Noon Wednesday December 18, 2014
=
NAME (Please print clearly):
SECTION:
I agree to complete this examination without unauthorized assistance from any person, materials or device.
Signat
550.420
Introduction to Probability
Fall 2014
Midterm Examination # 3
Wednesday December 3, 2014
=
NAME (Please print clearly):
SECTION:
I agree to complete this examination without unauthorized assistance from any person, materials, or device.
Signature:
550.413: Applied Statistics and Data Analysis
Lecture 3: Diagnostics and Remedial Measures
Applied Mathematics and Statistic, Johns Hopkins University
Gauss-Markov conditions
Theorem
Under the regression model Yi = 0 + 1 Xi + i where E[i ] = 0,
Var[i ] =
550.413: Applied Statistics and Data Analysis
Lecture 2: Inferences in Regression and Correlation Analysis
Department of Applied Mathematics and Statistic
the Johns Hopkins University
Fall 2016
Normally distributed random variables
A random variable X is
550.413 Assignment 1
Fall 2016
Instruction: This assignment consists of 4 problems. The assignment is
due on Wednesday, September 28, 2016 at 3pm, in class. If you cannot
make it to class, please leave the assignment under the door at Whitehead 306E
and e
550.413 Assignment 1
Fall 2016
Instruction: This assignment consists of 4 problems. The assignment is due on
Wednesday, September 28, 2016 at 3pm, in class. If you cannot make it to
class, please leave the assignment under the door at Whitehead 306E and e
550.413: Applied Statistics and Data Analysis
Lecture 1: Introduction to Simple Regression
Department of Applied Mathematics and Statistics
Johns Hopkins University
Fall 2016
Course syllabus and administrative details
Instructor: Minh Tang ([email protected]
550.413: Applied Statistics and Data Analysis
Lecture 5: Introduction to matrices
Applied Mathematics and Statistic, Johns Hopkins University
Hitchhiker guide to basic results in matrices theory
Remark
Dont panic!
Additional resources on matrices and thei
550.413: Applied Statistics and Data Analysis
Lecture 4: Simultaneous inferences and other topics
Applied Mathematics and Statistic, Johns Hopkins University
Bonferroni correction
The Bonferroni correction is a general procedure for simultaneous
inference
gupfm'b X is a C’owhv’wm T‘—\/\ Low/Irma
“(Ike/(4 4304‘ ONES POKSRLLL VﬂLlA/w dc X) Say) 06 ant£ OHM; L70
vii—L
130 g ><s mm = j mm a: mom WLM Lawn.
[VOW ag 14—90]
BM— I‘J-‘goes h (9 {A yvol‘ a Wavy -J«Lm+
pg 5 X am) /
L —9 OCCx) a:
m~u _ - -' “h-_ _
550.413: Applied Statistics and Data Analysis
Lecture 4: Simultaneous inferences and other topics
Applied Mathematics and Statistic, Johns Hopkins University
Joint estimation of
0
and
1
We often times want to have a joint condence region for multiple
para
550.413: Applied Statistics and Data Analysis
Lecture 3: Diagnostics and Remedial Measures
Applied Mathematics and Statistic, Johns Hopkins University
Gauss-Markov conditions
Theorem
Under the regression model Yi = 0 + 1 Xi + i where E[i ] = 0,
Var[i ] =
550.413: Applied Statistics and Data Analysis
Lecture 2: Inferences in Regression and Correlation Analysis
Applied Mathematics and Statistic, the Johns Hopkins University
Normally distributed random variables
A random variable X is said to have the normal
550.413: Applied Statistics and Data Analysis
Lecture 1: Introduction to Regression
Department of Applied Mathematics and Statistic
Johns Hopkins University
Course syllabus and administrative details
Instructor: Minh Tang @ Whitehead 306E
Email: [email protected]