Failure Time of System due to the Hot Electron Effect

Similar documents
EXST Regression Techniques Page 1. We can also test the hypothesis H :" œ 0 versus H :"

Lecture 1 Linear Regression with One Predictor Variable.p2

Chapter 8 (More on Assumptions for the Simple Linear Regression)

Multicollinearity Exercise

Lecture 11: Simple Linear Regression

3 Variables: Cyberloafing Conscientiousness Age

PubH 7405: REGRESSION ANALYSIS SLR: DIAGNOSTICS & REMEDIES

Handout 1: Predicting GPA from SAT

Overview Scatter Plot Example

data proc sort proc corr run proc reg run proc glm run proc glm run proc glm run proc reg CONMAIN CONINT run proc reg DUMMAIN DUMINT run proc reg

EXST7015: Estimating tree weights from other morphometric variables Raw data print

Effect of Centering and Standardization in Moderation Analysis

SPECIAL TOPICS IN REGRESSION ANALYSIS

Lecture notes on Regression & SAS example demonstration

unadjusted model for baseline cholesterol 22:31 Monday, April 19,

IES 612/STA 4-573/STA Winter 2008 Week 1--IES 612-STA STA doc

SAS Procedures Inference about the Line ffl model statement in proc reg has many options ffl To construct confidence intervals use alpha=, clm, cli, c

Table 1: Fish Biomass data set on 26 streams

Stat 302 Statistical Software and Its Applications SAS: Simple Linear Regression

This is a Randomized Block Design (RBD) with a single factor treatment arrangement (2 levels) which are fixed.

In Class Review Exercises Vartanian: SW 540

5.3 Three-Stage Nested Design Example

General Linear Model (Chapter 4)

Outline. Review regression diagnostics Remedial measures Weighted regression Ridge regression Robust regression Bootstrapping

ssh tap sas913, sas

Topic 14: Inference in Multiple Regression

Chapter 1 Linear Regression with One Predictor

T-test: means of Spock's judge versus all other judges 1 12:10 Wednesday, January 5, judge1 N Mean Std Dev Std Err Minimum Maximum

Lecture 3: Inference in SLR

Statistics 5100 Spring 2018 Exam 1

COMPREHENSIVE WRITTEN EXAMINATION, PAPER III FRIDAY AUGUST 26, 2005, 9:00 A.M. 1:00 P.M. STATISTICS 174 QUESTION

Lab # 11: Correlation and Model Fitting

Chapter 8 Quantitative and Qualitative Predictors

Comparison of a Population Means

STOR 455 STATISTICAL METHODS I

1) Answer the following questions as true (T) or false (F) by circling the appropriate letter.

ANALYSES OF NCGS DATA FOR ALCOHOL STATUS CATEGORIES 1 22:46 Sunday, March 2, 2003

Chapter 11 : State SAT scores for 1982 Data Listing

****Lab 4, Feb 4: EDA and OLS and WLS

dm'log;clear;output;clear'; options ps=512 ls=99 nocenter nodate nonumber nolabel FORMCHAR=" = -/\<>*"; ODS LISTING;

Odor attraction CRD Page 1

ST505/S697R: Fall Homework 2 Solution.

Week 7.1--IES 612-STA STA doc

Measurement Error. Often a data set will contain imperfect measures of the data we would ideally like.

Fixed and Random Effects Models: Vartanian, SW 683

Gregory Carey, 1998 Regression & Path Analysis - 1 MULTIPLE REGRESSION AND PATH ANALYSIS

Correlation and Simple Linear Regression

4 Instrumental Variables Single endogenous variable One continuous instrument. 2

MEMORIAL UNIVERSITY OF NEWFOUNDLAND DEPARTMENT OF MATHEMATICS AND STATISTICS FINAL EXAM - STATISTICS FALL 1999

Measuring relationships among multiple responses

Chapter 2 Inferences in Simple Linear Regression

Statistics for exp. medical researchers Regression and Correlation

Handout 12. Endogeneity & Simultaneous Equation Models

4 Instrumental Variables Single endogenous variable One continuous instrument. 2

Handout 11: Measurement Error

using the beginning of all regression models

Lecture 3. Experiments with a Single Factor: ANOVA Montgomery 3-1 through 3-3

Booklet of Code and Output for STAC32 Final Exam

Statistical Modelling in Stata 5: Linear Models

Exercices for Applied Econometrics A

7.3 Ridge Analysis of the Response Surface

SAS Commands. General Plan. Output. Construct scatterplot / interaction plot. Run full model

Lecture#12. Instrumental variables regression Causal parameters III

Topic 18: Model Selection and Diagnostics

LONGITUDINAL DATA ANALYSIS Homework I, 2005 SOLUTION. A = ( 2) = 36; B = ( 4) = 94. Therefore A B = 36 ( 94) = 3384.

The program for the following sections follows.

Lecture 4. Checking Model Adequacy

^l,2-, 3. f 2_ ' >v., 4. b/-e. of- fee. -^3-5- I70(fa = )&*>$)

Booklet of Code and Output for STAC32 Final Exam

2.1. Consider the following production function, known in the literature as the transcendental production function (TPF).

Paper: ST-161. Techniques for Evidence-Based Decision Making Using SAS Ian Stockwell, The Hilltop UMBC, Baltimore, MD

Model Selection Procedures

Problem Set 10: Panel Data

Answers: Problem Set 9. Dynamic Models

Correlation and the Analysis of Variance Approach to Simple Linear Regression

Specification Error: Omitted and Extraneous Variables

Problem Set #5-Key Sonoma State University Dr. Cuellar Economics 317- Introduction to Econometrics

STATISTICS 110/201 PRACTICE FINAL EXAM

sociology sociology Scatterplots Quantitative Research Methods: Introduction to correlation and regression Age vs Income

Lecture 3 Linear random intercept models

[The following data appear in Wooldridge Q2.3.] The table below contains the ACT score and college GPA for eight college students.

STATISTICS 479 Exam II (100 points)

Lecture 11 Multiple Linear Regression

Residuals from regression on original data 1

. *DEFINITIONS OF ARTIFICIAL DATA SET. mat m=(12,20,0) /*matrix of means of RHS vars: edu, exp, error*/

Lecture 3. Experiments with a Single Factor: ANOVA Montgomery 3.1 through 3.3

Chapter 6 Multiple Regression

Quantitative Methods Final Exam (2017/1)

SplineLinear.doc 1 # 9 Last save: Saturday, 9. December 2006

A Note on Decomposingthe Difference of Two Ratios. Daniel R. Corro

Interaction effects between continuous variables (Optional)

Statistics 512: Applied Linear Models. Topic 1

Practice 2SLS with Artificial Data Part 1

Regression without measurement error using proc calis

SAS code for Levene s Test and Weighted Least Squares DM LOG; CLEAR; OUT; CLEAR; ; ODS GRAPHICS ON; OPTIONS NODATE NONUMBER;

At this point, if you ve done everything correctly, you should have data that looks something like:

sociology 362 regression

Lab 07 Introduction to Econometrics

Chap 10: Diagnostics, p384

Lecture 8: Instrumental Variables Estimation

Transcription:

of System due to the Hot Electron Effect 1 * exresist; 2 option ls=120 ps=75 nocenter nodate; 3 title of System due to the Hot Electron Effect ; 4 * TIME = failure time (hours) of a system due to drift caused by hot electrons; 5 * RESIST = resistance (micro ohms) of a key resistor; 6 7 data CIRCUIT; input RESIST TIME @@; RESIST2=RESIST*RESIST; 8 RESIST3=RESIST*RESIST2; RESIST4=RESIST2*RESIST2; 9 R1=RESIST-9.889; R2=R1*R1; R3=R1*R2; R4=R2*R2; 10 label TIME= RESIST= Resistance RESIST2= Squared Resistance 11 12 cards; R= Centered Resistance ; 13 10.54 4.62 10.18 4.99 10.59 4.48 14 10.16 5.03 9.50 4.81 10.10 4.98 15 16 9.62 9.92 4.86 5.04 10.08 10.02 4.94 5.05 9.92 10.19 4.79 5.01 17 18 9.57 9.15 4.97 4.14 9.74 9.59 4.96 4.73 9.80 9.81 4.90 4.90 19 9.55 4.74 9.66 4.99 10.05 5.15 20 21 9.68 9.93 4.87 4.97 10.61 9.89 4.71 4.95 10.31 9.81 4.88 5.17 22 9.20 4.35 10.10 4.94 9.40 4.61 23 24 10. proc print; var TIME RESIST r1; 25 proc plot; plot TIME*RESIST; 26 proc reg; model TIME=RESIST RESIST2/r cli clm; 27 output out=out1 r=resid p=pred; 28 label RESID= Residuals PRED= Predicted Values ; 29 30 proc plot; plot RESID*RESIST/vref=0; 31 proc rank normal=blom; var RESID; ranks QNORM; 32 proc plot; plot RESID*QNORM; 33 label QNORM= Normal Quantiles ; 34 35 proc reg; model TIME=RESIST; 36 proc reg; model TIME=RESIST RESIST2 RESIST3; 37 proc reg; model time=resist resist2-resist4; 38 39 proc corr; var resist resist2-resist4; 40 proc corr; var r1-r4; 41 42 proc reg; model time=r1-r3; 43 proc reg; model time=r1-r4;

of System due to the Hot Electron Effect OBS TIME RESIST R1 1 4.62 10.54 0.651 2 4.99 10.18 0.291 3 4.48 10.59 0.701 4 5.03 10.16 0.271 5 4.81 9.50-0.389 6 4.98 10.10 0.211 7 4.86 9.62-0.269 8 4.94 10.08 0.191 9 4.79 9.92 0.031 10 5.04 9.92 0.031 11 5.05 10.02 0.131 12 5.01 10.19 0.301 13 4.97 9.57-0.319 14 4.96 9.74-0.149 15 4.90 9.80-0.089 16 4.14 9.15-0.739 17 4.73 9.59-0.299 18 19 4.90 4.74 9.81 9.55-0.079-0.339 20 21 4.99 5.15 9.66 10.05-0.229 0.161 22 4.87 9.68-0.209 23 24 4.71 4.88 10.61 10.31 0.721 0.421 25 4.97 9.93 0.041 26 4.95 9.89 0.001 27 5.17 9.81-0.079 28 29 4.35 4.94 9.20 10.10-0.689 0.211 30 4.61 9.40-0.489 2

of System due to the Hot Electron Effect 2 Plot of TIME*RESIST. Legend: A = 1 obs, B = 2 obs, etc. 5.2 + A A 5.1 + A A A A 5.0 + A A A A A A A A A 4.9 + B A A A A 4.8 + A F a i A A l A u 4.7 + r e T A A i 4.6 + m e 4.5 + A 4.4 + A 4.3 + 4.2 + A 4.1 + ---+------+------+------+------+------+------+------+------+------+------+------+------+------+------+------+-- 9.15 9.25 9.35 9.45 9.55 9.65 9.75 9.85 9.95 10.05 10.15 10.25 10.35 10.45 10.55 10.65 NOTE: 1 obs had missing values. Resistance 3

of System due to the Hot Electron Effect 3 Model: MODEL1 Model 2 1.27669 0.63834 74.232 0.0001 Error 27 0.23218 0.00860 Root MSE 0.09273 R-square 0.8461 Dep Mean 4.85100 Adj R-sq 0.8347 C.V. 1.91161 Variable DF Estimate Error Parameter=0 Prob > T Label INTERCEP 1-109.234255 9.75925927-11.193 0.0001 Intercept RESIST 1 22.882352 1.97165807 11.606 0.0001 Resistance RESIST2 1-1.145765 0.09950453-11.515 0.0001 Squared Resistance of System due to the Hot Electron Effect 4 Dep Var Predict Std Err Lower95% Upper95% Lower95% Upper95% Std Err Student Obs TIME Value Predict Mean Mean Predict Predict Residual Residual Residual 1 4.6200 4.6609 0.044 4.5706 4.7511 4.4503 4.8715-0.0409 0.082-0.501 2 4.9900 4.9697 0.022 4.9237 5.0157 4.7740 5.1655 0.0203 0.090 0.225 3 4.4800 4.5945 0.050 4.4923 4.6967 4.3785 4.8105-0.1145 0.078-1.464 4 5.0300 4.9782 0.022 4.9327 5.0236 4.7825 5.1738 0.0518 0.090 0.576 5 4.8100 4.7428 0.025 4.6915 4.7941 4.5457 4.9399 0.0672 0.089 0.753 6 4.9800 4.9980 0.022 4.9535 5.0425 4.8026 5.1934-0.0180 0.090-0.200 7 4.8600 4.8598 0.022 4.8154 4.9043 4.6644 5.0552 0.000164 0.090 0.002 8 4.9400 5.0028 0.022 4.9584 5.0472 4.8074 5.1982-0.0628 0.090-0.696 9 4.7900 5.0081 0.021 4.9640 5.0521 4.8128 5.2034-0.2181 0.090-2.417 10 5.0400 5.0081 0.021 4.9640 5.0521 4.8128 5.2034 0.0319 0.090 0.354 11 5.0500 5.0116 0.022 4.9674 5.0558 4.8163 5.2070 0.0384 0.090 0.425 12 5.0100 4.9651 0.023 4.9189 5.0114 4.7693 5.1610 0.0449 0.090 0.499 13 4.9700 4.8151 0.023 4.7687 4.8615 4.6192 5.0109 0.1549 0.090 1.723 14 4.9600 4.9439 0.021 4.9006 4.9871 4.7488 5.1390 0.0161 0.090 0.179 15 4.9000 4.9735 0.021 4.9300 5.0170 4.7783 5.1687-0.0735 0.090-0.814 16 4.1400 4.2130 0.058 4.0939 4.3320 3.9885 4.4374-0.0730 0.072-1.008 17 4.7300 4.8337 0.022 4.7882 4.8792 4.6380 5.0293-0.1037 0.090-1.151 18 4.9000 4.9777 0.021 4.9341 5.0212 4.7825 5.1729-0.0777 0.090-0.860 19 4.7400 4.7956 0.023 4.7480 4.8431 4.5995 4.9917-0.0556 0.090-0.619 20 4.9900 4.8915 0.021 4.8479 4.9352 4.6963 5.0867 0.0985 0.090 1.091 21 5.1500 5.0083 0.022 4.9640 5.0525 4.8129 5.2036 0.1417 0.090 1.572 22 4.8700 4.9060 0.021 4.8626 4.9494 4.7108 5.1011-0.0360 0.090-0.399 23 4.7100 4.5663 0.052 4.4590 4.6737 4.3479 4.7848 0.1437 0.077 1.876 24 4.8800 4.8924 0.026 4.8389 4.9460 4.6948 5.0901-0.0124 0.089-0.140 25 4.9700 5.0095 0.021 4.9654 5.0536 4.8141 5.2048-0.0395 0.090-0.437 26 4.9500 5.0025 0.021 4.9585 5.0465 4.8072 5.1978-0.0525 0.090-0.582 27 5.1700 4.9777 0.021 4.9341 5.0212 4.7825 5.1729 0.1923 0.090 2.131 28 4.3500 4.3058 0.051 4.2006 4.4111 4.0884 4.5233 0.0442 0.077 0.572 29 4.9400 4.9980 0.022 4.9535 5.0425 4.8026 5.1934-0.0580 0.090-0.643 30 4.6100 4.6201 0.031 4.5567 4.6834 4.4195 4.8206-0.0101 0.087-0.115 31. 5.0128 0.022 4.9686 5.0569 4.8174 5.2081... Sum of Residuals 4.38E-13 Sum of Squared Residuals 0.2322 Predicted Resid SS (Press) 0.3037 4

of System due to the Hot Electron Effect 5 Plot of RESID*RESIST. Legend: A = 1 obs, B = 2 obs, etc. 0.20 + A A 0.15 + A A 0.10 + A A 0.05 + A A A A A A A R e 0.00 +-----------------------------------A-------------------------------------------------------------------------- s A i A d A u a A l A A s -0.05 + A A A A A A A -0.10 + A A -0.15 + -0.20 + A -0.25 + ---+------+------+------+------+------+------+------+------+------+------+------+------+------+------+------+-- 9.15 9.25 9.35 9.45 9.55 9.65 9.75 9.85 9.95 10.05 10.15 10.25 10.35 10.45 10.55 10.65 NOTE: 1 obs had missing values. Resistance 5

of System due to the Hot Electron Effect 6 Plot of RESID*QNORM. Legend: A = 1 obs, B = 2 obs, etc. 0.20 + A A 0.15 + A A 0.10 + A A 0.05 + A A A A A A A R e 0.00 + A s A i A d A u a A l A A s -0.05 + A A A A A A -0.10 + A A -0.15 + A -0.20 + A -0.25 + -+----------+----------+----------+----------+----------+----------+----------+----------+----------+----------+ -2.5-2.0-1.5-1.0-0.5 0.0 0.5 1.0 1.5 2.0 2.5 NOTE: 1 obs had missing values. Normal Quantiles 6

of System due to the Hot Electron Effect 7 Model 1 0.13652 0.13652 2.785 0.1063 Error 28 1.37235 0.04901 Root MSE 0.22139 R-square 0.0905 Dep Mean 4.85100 Adj R-sq 0.0580 Variable DF Estimate Error Parameter=0 Prob > T Label INTERCEP 1 3.015242 1.10067546 2.739 0.0106 Intercept RESIST 1 0.185636 0.11122794 1.669 0.1063 Resistance Model 3 1.27843 0.42614 48.081 0.0001 Error 26 0.23044 0.00886 Root MSE 0.09414 R-square 0.8473 Dep Mean 4.85100 Adj R-sq 0.8297 Variable DF Estimate Error Parameter=0 Prob > T Label INTERCEP 1-216.439192 241.98353889-0.894 0.3793 Intercept RESIST 1 55.462179 73.50494647 0.755 0.4573 Resistance RESIST2 1-4.442173 7.43511582-0.597 0.5554 Squared Resistance RESIST3 1 0.111043 0.25043680 0.443 0.6611 Model 3 1.27843 0.42614 48.081 0.0001 Error 26 0.23044 0.00886 Root MSE 0.09414 R-square 0.8473 Dep Mean 4.85100 Adj R-sq 0.8297 NOTE: Model is not full rank. Least-squares solutions for the parameters are not unique. Some statistics will be misleading. A reported DF of 0 or B means that the estimate is biased. The following parameters have been set to 0, since the variables are a linear combination of other variables as shown. RESIST4 = -9462 * INTERCEP +3843 * RESIST -584.6723 * RESIST2 +39.5032 * RESIST3 Variable DF Estimate Error Parameter=0 Prob > T Label INTERCEP B -216.439192 241.98353889-0.894 0.3793 Intercept RESIST B 55.462179 73.50494647 0.755 0.4573 Resistance RESIST2 B -4.442173 7.43511582-0.597 0.5554 Squared Resistance RESIST3 B 0.111043 0.25043680 0.443 0.6611 RESIST4 0 0... 7

of System due to the Hot Electron Effect 10 Correlation Analysis 4 VAR Variables: RESIST RESIST2 RESIST3 RESIST4 Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label RESIST 31 9.892581 0.363941 306.670000 9.150000 10.610000 Resistance RESIST2 31 97.991332 7.210220 3037.731300 83.722500 112.572100 Squared Resistance RESIST3 31 971.925885 107.321884 30130 766.060875 1194.389981 RESIST4 31 9652.611463 1422.428945 299231 7009.457006 12672 Pearson Correlation Coefficients / Prob > R under Ho: Rho=0 / N = 31 RESIST RESIST2 RESIST3 RESIST4 RESIST 1.00000 0.99972 0.99887 0.99748 Resistance 0.0 0.0001 0.0001 0.0001 RESIST2 0.99972 1.00000 0.99972 0.99888 Squared Resistance 0.0001 0.0 0.0001 0.0001 RESIST3 0.99887 0.99972 1.00000 0.99972 0.0001 0.0001 0.0 0.0001 RESIST4 0.99748 0.99888 0.99972 1.00000 0.0001 0.0001 0.0001 0.0 of System due to the Hot Electron Effect 11 Correlation Analysis 4 VAR Variables: R1 R2 R3 R4 Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum R1 31 0.003581 0.363941 0.111000-0.739000 0.721000 R2 31 0.128193 0.171877 3.973991 0.000001000 0.546121 R3 31 0.004034 0.145284 0.125066-0.403583 0.374805 R4 31 0.045022 0.090521 1.395690 1.000006E-12 0.298248 Pearson Correlation Coefficients / Prob > R under Ho: Rho=0 / N = 31 R1 R2 R3 R4 R1 1.00000 0.05906 0.87959 0.07214 0.0 0.7523 0.0001 0.6997 R2 0.05906 1.00000 0.08045 0.97441 0.7523 0.0 0.6670 0.0001 R3 0.87959 0.08045 1.00000 0.07087 0.0001 0.6670 0.0 0.7048 R4 0.07214 0.97441 0.07087 1.00000 0.6997 0.0001 0.7048 0.0 8

of System due to the Hot Electron Effect 12 Model: MODEL1 Model 3 1.27843 0.42614 48.081 0.0001 Error 26 0.23044 0.00886 Root MSE 0.09414 R-square 0.8473 Dep Mean 4.85100 Adj R-sq 0.8297 C.V. 1.94071 Variable DF Estimate Error Parameter=0 Prob > T Label INTERCEP 1 5.002123 0.02176143 229.862 0.0001 Intercept R1 1 0.182361 0.10001858 1.823 0.0798 R2 1-1.147858 0.10112907-11.350 0.0001 R3 1 0.111043 0.25043667 0.443 0.6611 of System due to the Hot Electron Effect 13 Model: MODEL1 Model 4 1.28114 0.32029 35.161 0.0001 Error 25 0.22773 0.00911 Root MSE 0.09544 R-square 0.8491 Dep Mean 4.85100 Adj R-sq 0.8249 C.V. 1.96747 Variable DF Estimate Error Parameter=0 Prob > T Label INTERCEP 1 4.991867 0.02898859 172.201 0.0001 Intercept R1 1 0.192802 0.10318903 1.868 0.0735 R2 1-0.900328 0.46527614-1.935 0.0644 R3 1 0.085698 0.25810789 0.332 0.7426 R4 1-0.479960 0.87999520-0.545 0.5903 9