How To Interpret Regression Analysis Results - P-Values and Coefficients - Minitab PDF

5/18/2015
HowtoInterpretRegressionAnalysisResults:PvaluesandCoefficients|Minitab
The Minitab Blog
http://blog.minitab.com
Data Analysis http://blog.minitab.com/blog/dataanalysis2

Quality Improvement http://blog.minitab.com/blog/qualityimprovement2
How
tohttp://blog.minitab.com/blog/projecttools2
Interpret Regression Analysis Results:
Pvalues
Project Tools
Minitab.com http://www.minitab.com
and Coefficients
Jim Frost http://blog.minitab.com/blog/adventuresinstatistics . 1 July, 2013
98
778
153
59 http://blog.minitab.com/blog/adventuresinstatistics/howtointerpretregressionanalysisresultspvaluesandcoefficients
Master Statistics
Anytime,
Anywhere
Quality Trainer teaches
you how to analyze
your data anytime you
are online.
Regression analysis generates an equation to describe the statistical relationship between

one or more predictor variables and the response variable. After you use Minitab Statistical
Software http://www.minitab.com/enus/products/minitab/ to fit a regression model, and
verify the fit by checking the residual plots http://blog.minitab.com/blog/adventuresin
statistics/whyyouneedtocheckyourresidualplotsforregressionanalysis, youll want to
interpret the results. In this post, Ill show you how to interpret the pvalues and coefficients
that appear in the output for linear regression analysis.
How Do I Interpret the PValues in Linear Regression

Analysis?
The pvalue for each term tests the null hypothesis that the coefficient is equal to zero no
effect. A low pvalue < 0.05 indicates that you can reject the null hypothesis. In other
Take the Tour!
http://www.minitab.com/products/quality
words, a predictor that has a low pvalue is likely to be a meaningful addition to your model
trainer/?
because changes in the predictor's value are related to changes in the response variable.
WT.ac=BlogQT
Conversely, a larger insignificant pvalue suggests that changes in the predictor are not
associated with changes in the response.
In the output below, we can see that the predictor variables of South and North are
significant because both of their pvalues are 0.000. However, the pvalue for East 0.092 is
greater than the common alpha level of 0.05, which indicates that it is not statistically
significant.
Typically, you use the coefficient pvalues to determine which terms to keep in the
regression model. In the model above, we should consider removing East.
http://blog.minitab.com/blog/adventuresinstatistics/howtointerpretregressionanalysisresultspvaluesandcoefficients
1/22
5/18/2015
How Do I Interpret the Regression Coefficients for Linear

Relationships?
Regression coefficients represent the mean change in the response variable for one unit of
change in the predictor variable while holding other predictors in the model constant. This
statistical control http://blog.minitab.com/blog/adventuresinstatistics/atributeto
regressionanalysis that regression provides is important because it isolates the role of one
variable from all of the others in the model.
The key to understanding the coefficients is to think of them as slopes, and theyre often
called slope coefficients. Ill illustrate this in the fitted line plot below, where Ill use a
persons height to model their weight. First, Minitabs session window output:
The fitted line plot shows the same regression results graphically.
The equation shows that the coefficient for height in meters is 106.5 kilograms. The
coefficient indicates that for every additional meter in height you can expect weight to
increase by an average of 106.5 kilograms.
The blue fitted line graphically shows the same information. If you move left or right along
the xaxis by an amount that represents a one meter change in height, the fitted line rises or
falls by 106.5 kilograms. However, these heights are from middleschool aged girls and
range from 1.3 m to 1.7 m. The relationship is only valid within this data range, so we would
not actually shift up or down the line by a full meter in this case.
2/22
5/18/2015
If the fitted line was flat a slope coefficient of zero, the expected value for weight would
not change no matter how far up and down the line you go. So, a low pvalue suggests that
the slope is not zero, which in turn suggests that changes in the predictor variable are
associated with changes in the response variable.
I used a fitted line plot because it really brings the math to life. However, fitted line plots can
only display the results from simple regression, which is one predictor variable and the
response. The concepts hold true for multiple linear regression, but I would need an extra
spatial dimension for each additional predictor to plot the results. That's hard to show with
today's technology!
How Do I Interpret the Regression Coefficients for

Curvilinear Relationships and Interaction Terms?
In the above example, height is a linear effect; the slope is constant, which indicates that the
effect is also constant along the entire fitted line. However, if your model requires
polynomial or interaction terms, the interpretation is a bit less intuitive.
As a refresher, polynomial terms model curvature in the data
http://blog.minitab.com/blog/adventuresinstatistics/curvefittingwithlinearand
nonlinearregression, while interaction terms indicate that the effect of one predictor
depends on the value of another predictor.
The next example uses a data set that requires a quadratic squared term to model the
curvature. In the output below, we see that the pvalues for both the linear and quadratic
terms are significant.
The residual plots not shown indicate a good fit, so we can proceed with the interpretation.
But, how do we interpret these coefficients? It really helps to graph it in a fitted line plot.
3/22
5/18/2015
You can see how the relationship between the machine setting and energy consumption
varies depending on where you start on the fitted line. For example, if you start at a machine
setting of 12 and increase the setting by 1, youd expect energy consumption to decrease.
However, if you start at 25, an increase of 1 should increase energy consumption. And if
youre around 20, energy consumption shouldnt change much at all.
A significant polynomial term can make the interpretation less intuitive because the effect of
changing the predictor varies depending on the value of that predictor. Similarly, a
significant interaction term indicates that the effect of the predictor varies depending on the
value of a different predictor.
Take extra care when you interpret a regression model that contains these types of terms.
You cant just look at the main effect linear term and understand what is happening!
Unfortunately, if you are performing multiple regression analysis, you won't be able to use a
fitted line plot to graphically interpret the results. This is where subject area knowledge is
extra valuable!
Particularly attentive readers may have noticed that I didnt tell you how to interpret the
constant http://blog.minitab.com/blog/adventuresinstatistics/regressionanalysishowto
interprettheconstantyintercept. Ill cover that in my next post!
Be sure to:
Check your residual plots so you can trust the results
http://blog.minitab.com/blog/adventuresinstatistics/whyyouneedtocheckyour
residualplotsforregressionanalysis
Assess the goodnessoffit and Rsquared http://blog.minitab.com/blog/adventures
instatistics/regressionanalysishowdoiinterpretrsquaredandassessthe
goodnessoffit
4/22
5/18/2015
If you're learning about regression, read my regression tutorial

http://blog.minitab.com/blog/adventuresinstatistics/regressionanalysistutorialand
examples!
You Might Also Like:

Regression Analysis: How to Interpret the Constant Y Intercept http://blog.minitab.com/blog/adventuresin
statistics/regressionanalysishowtointerprettheconstantyintercept
Regression Analysis: How Do I Interpret Rsquared and Assess the GoodnessofFit?
http://blog.minitab.com/blog/adventuresinstatistics/regressionanalysishowdoiinterpretrsquaredandassessthe
goodnessoffit
Why Are There No P Values for the Variables in Nonlinear Regression? http://blog.minitab.com/blog/adventuresin
statistics/whyaretherenopvaluesforthevariablesinnonlinearregression
Regression Analysis Tutorial and Examples http://blog.minitab.com/blog/adventuresinstatistics/regressionanalysis
tutorialandexamples
Comments
Name: Lovemore Friday, January 24, 2014
That's sounds great but for me I am finding difficult how do I instigate a six sigma project in a medical laboratory using so
of the Minitab tools
Name: Henry Mwangi Thursday, February 20, 2014

Thank you for an elaborate explanation on the interpreting reg coefficients and mostly the pvalue.
Name: Deeps Dee Thursday, March 27, 2014

It has been useful for my thesis whereby I've been struggling to interpret my results :s
Thank you for the explanation.
Name: taiwo lucas Wednesday, April 2, 2014

Thank you very much the explanation really help me in my thesis.God bless you.
Name: O.Jobi Saturday, May 10, 2014

This is very helpful information for my dissertation page 4&5.
Name: yashika Tuesday, May 13, 2014

really i was confused and you clear this concept of regression coefficient. very good explanation.
can you do this with ttest explanation also?
5/22
5/18/2015
Name: omid Saturday, June 7, 2014

hi dear,
I am doing a censored least absolute deviation model using STATA, when I got output there was a column indicated with
"Bias" , does it mean Pvalue ?
Name: Jim Frost Monday, June 9, 2014

Hi Omid,
Thanks for your question. I can't really offer guidance about using Stata. However, bias and Pvalue are not synonymous, so
that's probably not what the output means.
I suspect it has to do with the censoring in your data. Regression with censored data can cause biased estimates because
you may be less likely to observe the response value for certain classes of observations. In other words, the model that fits
the observed responses may not provide an unbiased fit for the censored observations.
Minitab can perform regression with censored data and can assume different distributions. In Minitab: Stat >
Reliability/Survival > Regression with Life Data.
You can try a free 30 day trial of Minitab 17 here:
http://it.minitab.com/enus/products/minitab/freetrial.aspx http://it.minitab.com/enus/products/minitab/freetrial.aspx
Thanks for writing!
Jim
Name: Mrv Yrd Wednesday, August 27, 2014

Hi Jim,
First of all Thank you for the useful information! I am little confused about p value and significance for regression. If our p
value is 0.02 for SLR can we say that regression analysis is statistically significant at 95% confidence level ? Or should we say
it is significant at 98%?
My second question is that if we are not given the p value for the variable and the constant for SLR, but the regression p
value is smaller than 0.05 , can we conclude the factor significantly affects the response ? Thank you in advance.
Name: Jim Frost Thursday, August 28, 2014

Hi,
Typically you choose the significance level before the study, and that's the level you cite after the analysis. For example, you
can state that the SLR is statistically significant at the the 0.05 level. Or for multiple regression, identify the variables that
are significant at that level e.g. 0.05. You typically don't change the significance level to match your pvalues.
However, I'd also report the exact pvalues as well. The exact pvalue is important in terms of understanding the liklihood
that your test drew the correct conclusions. I cover that in this post:
http://blog.minitab.com/blog/adventuresinstatistics/fiveguidelinesforusingpvalues
http://blog.minitab.com/blog/adventuresinstatistics/fiveguidelinesforusingpvalues
For your second question. Yes, in a simple linear regression model Y = a + bX, the regression pvalue in the ANOVA is for
a test of the hypothesis that the linear coefficient is zero.
6/22
5/18/2015
Thanks for reading!

Jim
49Comments
TheMinitabBlog
Recommend
Share
Login
SortbyOldest
Jointhediscussion
Joel 6monthsago
Hello,
Ifittedthemodely=a+bX1+cX2+dX1.X2+e(X1)^2+f(X2)^2onadatasetbutIhavesomeproblemsin
interpretingthepvaluesofthecoefficients.IfIusenormalizedvaluesforX1andX2(smallestvalue:1,
largestvalue:+1)andIperformaregressionIgetdifferentpvaluesforthecoefficientsa,bandc(notd,e
andf)comparedtottherealvalues.Infactformydatasetp<0.05forX2ifInormalizemydatabut>0.05for
therealvalues.SoIguessnormalizationistobedonealwaystoanalyzedata?
Thanksinadvance.
Jol
1
Reply Share
inez 6monthsago
Inmylinearregressionresults,whatdothetvaluesmean?caniputthemintableofresults?
Reply Share
JimFrostAtMinitab
Mod >inez 6monthsago
HiInez!Thanksforwritingwiththeexcellentquestion!
Thetvalueisastatisticthatmeasurestheratiobetweenthecoefficientanditsstandarderror.
Minitabusesthetvaluetocalculatethepvalue,whichyouusetomakeadecisionaboutthe
statisticalsignificanceofthetermsandmodel.
Asufficientlylargeratioindicatesthatthecoefficientestimateisbothlargeandpreciseenoughtobe
significantlydifferentfromzero.Conversely,asmallratioindicatesthatthecoefficientestimateistoo
smallortooimprecisetobecertainthatthetermhasaneffectontheresponse.
Youcanusethetvaluetodeterminewhethertorejectthenullhypothesis.However,thepvalueis
usedmoreoftenbecauseitiseasiertointerpret.
Unlessyouhaveaspecialneedtoincludeit,Iwouldnotincludeitinyourresults.
Jim
Reply Share
Cain 5monthsago
HowcanItellthelevelofsignificancefromanoutput?IhaveanexamusingminitabandI'mnotsure
7/22
5/18/2015
Reply Share
JimFrostAtMinitab
Mod >Cain 5monthsago
Hi,thatsoundslikeatrickquestiontome.Thesignificancelevel(alpha)issomethingthatyoushould
choosebeforeyouperformyourstudy.Afteryouperformtheanalysis,youcomparethepvaluesin
theoutputtoyoursignificancelevel.
Jim
Reply Share
WDC123 5monthsago
HiJim,IfIreducethemodelbytakingouttermswithpvalueslessthank0.05andthennoticethatRsquared
hasalsoreducedhowdoIexplain.ShouldIconsiderleavinginsometerms?
Reply Share
JimFrostAtMinitab
Mod >WDC123 4monthsago
Hi,typicallyyouconsiderremovingpredictorsfromthemodelifthepvalueisgreaterthanyour
significancelevel.I'llassumethatiswhatyoumeanttotype!:)
It'sfairlytypicalfortheRsquaredtodeclineasyouremovepredictors,evenwhenthosepredictors
arenotsignificant.
Hereareacoupleofsuggestions:
*UseadjustedRsquaredtocomparemodelswithdifferentnumbersofterms.
*Don'tchoosethemodelbasedsolelyonthehighestRsquaredbecausethatcanleadyouastray.
*Useyourexpertise,theory,andcommonsenseratherthanrelyingsolelyonsimplisticmodel
selectionrules.
Foryourcase,don'tfeellikeyoushouldincludethoseinsignificantpredictorsjusttogetthehigherR
squared.However,youcanconsiderincludingthemiftheorysuggeststhattheybelonginthemodel.
Ingeneral,youshouldalreadyhaveanideaofwhattheimportantvariablesarealongwiththeir
relationships,coefficientsigns,andeffectmagnitudesbasedonpreviousresearch.
There'snotalwaysaclearansweronwhichpredictorsyoushouldincludeinyourmodel.Useboth
thestatisticaloutputandtheoretical/subjectareaconsiderationstohelpyoudecide.
Thanksforwritingwiththegreatquestion!Selectingthecorrectmodelhasalwaysbeenavery
interestingsubjectforme!
Jim
Reply Share
Ronja 4monthsago
Hello,
myquestionisquitesimilartotheothers:
inordertodevelopaforecastIwanttousemultipleregression.Itriedvariousindependentvariablesthat
wouldallmakesense(meaningtheyallmayhaveanimpactontheforecast)togainthebestsuitedequation
fortheforecast,butIfinditdifficulttochoosetherightsetofvariables.Withtheonesetofindependent
variables,mypvaluesarehigherthan0,05(theyare0,12)howevertheRsquaredishighestwith0,9904.
Takingouttermswithpvalueshigherthan0.05won'tworksincethentherewon'tbeanyleft.Withtheother
set,myRsquaredisjust0.8473howeverthepvaluesarelessthan0,05.Howdoyouselecttherightset?
DoyouweighthepvaluehigherortheRsquaredoristhereanothertermIshouldconsiderformy
8/22
5/18/2015
DoyouweighthepvaluehigherortheRsquaredoristhereanothertermIshouldconsiderformy
selection?
Thankyouverymuchinadvance!!!
Ronja
Reply Share
JimFrostAtMinitab
Mod >Ronja 4monthsago
HiRonja,
Selectingthecorrectmodelcanbeaverydifficultprocessinsomecases.Readmyresponsetothe
commentdirectlyaboveyours(toWDC123)becauseitappliestoyourcaseaswell.
Specifically,don'tfeellikeyoumustgetthehigherRsquaredbecauseit'spossibletohaveanR
squaredthatistoohighandcauseproblems.YourRsquaredof0.99maybetoohighandcould
indicatethatyou'reoverfittingthemodel.Also,youshoulduseadjustedRsquaredtocompare
modelswithdifferentnumbersofpredictorsratherthanRsquared.
IsuggestthatyoureadmyblogpostaboutadjustedRsquared,whichcoversalloftheabovepoints.
AsforpvaluesversusadjustedRsquaredvalues,researchhasshownthatusingpvaluesina
stepwisemannergenerallyworksbetterthanusingadjustedRsquaredtopickthecorrectmodel.
However,usinganysimplemodelselectionprocedurelikethatgenerallydoesnotpickthecorrect
model.I'vewrittenanotherpostaboutthisissuewhereIcomparestepwisetobestsubsets
regression.
Theimplicationsofthesefindingsareprofoundevenifyou'renotusingeitheroftheseautomated
methods.Thefindingsshowthatchoosingthecorrectmodelisasmuchascienceasitisanart.The
seemore
Reply Share
Fiachra 4monthsago
Hi,AfterrunningmyregressionIendedupwithpvalueslike6.9345E05.WhatdoesthisEmeanandhow
doIworkoutthePvaluethanks.
Reply Share
JimFrostAtMinitab
Mod >Fiachra 4monthsago
Thatiscalledscientificnotationandisusedtowriteverylargeandverysmallnumbers.Itworksby
shiftingthedecimalpointleftorrightbythenumberofplacesindicatedaftertheE,whichstandsfor
exponent.
The05indicatesthatyouneedtotakethe6.9345andshiftthedecimalpointtotheleftby5places.
So,yourpvalueis0.000069345.That'saverylowvaluesoitisverysignificant!
Jim
Reply Share
Fiachra>JimFrostAtMinitab 4monthsago
Thanksamillion!myheadwaswreckedthinkingitwassomethingmuchmorecomplex.Ido
haveoneotherquestionhoweverinarecentmcqIwasgivenaregressionoutputbasedon
salary=b1+b2(Rank).(Rankbeingthequalityoftheindividualsuniversity,thebestwas
awardedarankof1andtheworstarankof142).Thecoefficientstheregressionproduced
fortheinterceptandrankwere56063and206.731respectively.Bothhadveryverylowp
9/22
5/18/2015
fortheinterceptandrankwere56063and206.731respectively.Bothhadveryverylowp
valuessotheyweresignificant.Thequestionwaswhatisthetrueeffectofaoneplace
increaseinuniversityrankingsonsalaries.TheanswerIgavewas206.731butthecorrect
answeristhatitcannotbedeterminedfromthesefigures.(Figuresbeingapictureofa
regressionoutputinexcel).Whyisthisthecorrectanswer?Ithoughtthiswouldhavebeen
exactlywhatthecoefficientintheregressionindicates.
Thanks.
1
Reply Share
JimFrostAtMinitab
Mod >Fiachra 4monthsago
Ifthequestionaskedyouspecifically,whatwasthe"true"effect,youhaveto
rememberthatregression,andotherstatisticaltechniques,canonlyprovidean
estimateofthetrueeffect.It'sgenerallyimpossibletoeverknowthetrueeffectitself
becauseyou'reworkingwithasampleofthepopulationratherthantheentire
population.
Instead,inferentialstatisticscanonlyprovideanestimateofthetrueeffectandgive
youaconfidenceintervalforarangeofvaluesthatislikelytocontainthetrueeffect.
Inregressionanalysis,thecoefficientsaretheparameterestimates.
Reply Share
sewnsew 4monthsago
ihavearegressionmodelhowdoIcalculatethechangeinpwhenItakeoutvariablesoraddvariableback
intoamodeltoseewhichhasthemostpredictivevalue?InthedataIhave,Ihaveachangeinp,butin
SPSS,Idon'tseeanythingthatshowsorrelatestothechangeinp,sowhenIrerunthedata,Idon'tknow
whattolookfororwhattointerpretasachangeinp.Thanks.
Reply Share
JimFrostAtMinitab
Mod >sewnsew 4monthsago
Hi,Ican'tspeaktowhatyouseeinothersoftwarepackages.Also,I'mnotsurewhichpyouare
referringto.
Youmaywanttolookattheadjustedsumsofsquaresintheoutput.Thisindicatestheuniqueportion
ofthetotalsumsofsquaresthateachtermexplainsregardlessoftheordertheywereenteredinthe
model.Ifyouwanttofindouthowmuchvariationeachpredictorvariableaccountsforinamodel,this
iswhatyouneed.
Jim
Reply Share
Scott 4monthsago
Isthereanywaytoset/holdaparticularregressionequationcoefficientataparticularvalue,andthen
performtheregressionanalysis?
Inmyexample,Iamanalyzingpsioutvaluebasedonanumberofinputs,IwanttoholdPsiIncoefficientat
1,andlettheothervariablesbeapartoftheregression.Hopethismakessense,:/
Thanks!
Reply Share
JimFrostAtMinitab
HiScott,
Mod >Scott 4monthsago
10/22
5/18/2015
HiScott,
That'saninterestingquestion.
Typically,you'refittingamodellikethis:
Y=B0+BX1+BX2+BX3...whereyouestimatetheBsfromthedata.
Youwanttofitthis:
Y=B0+1X1+BX2+BX3...wherethefirstcoefficientis1.
WhatyoucantrydoingismovingthetermwiththefixedcoefficientovertotheYsideoftheequation:
Y1X1=B0+BX2+BX3
You'dhavetocreateanewcolumnofresponsedatawhereyoutaketheoriginalmeasureand
subtractoutthe1X1.Inyourcase,you'dtaketheoutputPSIandsubtracttheinputPSIforeach
observationandusethenewlycalculatedvaluesastheresponse.Then,includetherestofthe
predictorsinthemodel.
You'dessentiallybelookingathowthepredictorsarerelatedtothechangeinPSIratherthanthe
absolutePSI,whichsoundspromisingifIunderstandyourscenariocorrectly.
Theestimatesfortheotherpredictorswouldbethevaluesifforcedthefirstpredictortoequal1.
You'dhavetobecarefulhowyouinterpretthemodelfitvalues.Forexample,Rsquaredindicates
howmuchvariationyouaccountforwiththenewresponsevariable.
Jim
Reply Share
sewnsew 4monthsago
InmyhomogeneoussubsetstheNisdifferentthantheNthatIgotwhenIranfrequencies.Why?Isthis
normal?
Reply Share
SharonEdgeWilkie 4monthsago
ThisisaPostHocquestion.WhyaretheNinmyhomogeneoussubsetsnotthesameastheNinmy
frequencycharts?
Reply Share
PatrickKajubili 4monthsago
Hi,
Iamstilljuniorinthefield.iwanttoknowifihaveF714andSig761inmyANOVA
tablewhatdoesthismean?Havingsiglikedoesshowmodelfit?
Reply Share
JimFrostAtMinitab
Mod >PatrickKajubili 4monthsago
HiPatrick,theFstatisticisatestoftheoverallsignificanceoftheregressionmodel.WhileR
squaredandadjustedRsquaredtellyoutheoveralldegreeofthefitforaregressionmodel,they
don'tprovideaformalhypothesistestfortheoverallfit.
That'swheretheFtestanditsassociatedpvaluecomesin.
ThenullhypothesisfortheFtestisthatallofthecoefficientsintheregressionmodelequalzero.Ifall
thecoefficientsequalzero,thisisequivalenttosayingthatthefittedvaluessimplyequalthemeanof
11/22
5/18/2015
theresponsevariable.Inotherwords,yourmodelpredictstheresponsenobetterthanusingthe
responsemean.
Thealternativehypothesisisthattheydon'tallequalzero.Or,thatyourmodeldoesprovidebetter
predictionsthanjustusingthemean.
Alowpvaluemeansthatyoucanrejectthenullandconcludethatyourmodelisbetterthanjust
usingthemeanandthatatleastonecoefficientdoesn'tequalzero.
You'llstillneedtochecktheresidualplotsbecausethistestwon'ttellyouwhetherthemodel
providesanadequate,unbiasedfit.InthebulletsneartheendofthispostIprovidealinktoablog
postIwroteaboutcheckingtheresidualplots.
Thanksforwriting!
Jim
Reply Share
Marija 3monthsago
Hello,Ineedyurhelpaboutmyexamquestion:
(i)
Estimatethefollowingregressions
PRICE=b1+allindependentvariables+ut
LnPRICE=b1+allindependentvariables+ut
Accordingtotherelevantcriteria,judgewhichoneisbetter.Continueworkingwiththebetterfromthetwo.
Fullyinterpret(statisticalandeconomicsignificance)theresultsofhedonichousepriceestimation.
Myquestion:Whicharethecriteriatodecidewhichisabetterregression?
Ihavecalcualtedthembothandherearetheresults(valuesonlyfromthevariableswithsig.<0.05:
PRICE=27978,841+(140,661)+372,079+(12080,847)+11032,510+
(5154,908)+7478,822=29485,84
LnPRICE=10,5030,003+0,0060,247+0,1240,075+0,1126=10,42026
Ineedaninfofromyouinordertocontinueinterpretingtheresultsbasedonthebetterregression
Thankyouverymuchinadvance
Reply Share
JimFrostAtMinitab
Mod >Marija 3monthsago
HiMarija,
InadditiontothefactthatIreallyshouldnotansweryourexamquestionforyou,Ireallycan'tanswer
thequestionwiththeinformationthatyouprovided.Thereisinsufficientinformationtobeableto
choose.But,Icangiveyousomegeneralguidelinesonhowtochoose.
Youshouldchecktheresidualplotsforbothmodels.Iftheplotslookgoodforonemodelbutnotthe
other,thatwillhelpyouchoose.
Youshouldalsolookatthecoefficientsforthepredictorsanddeterminewhethertheymatchtheory.
Forexample,ifonemodelsuggeststhatagoodcharacteristiclowerstheprice(negativecoefficient),
youshouldseriouslyquestionthatmodel.
Thosearethetypesofthingsyouneedtoassesstodeterminewhichmodelisbetter.Irecently
wroteablogpostabouthowtochoosethebestregressionmodel.Ithinkthatwillhavealotofhelpful
12/22
5/18/2015
wroteablogpostabouthowtochoosethebestregressionmodel.Ithinkthatwillhavealotofhelpful
informationforyou!
Goodluckwithyourtest!
Jim
Reply Share
wuyr 3monthsago
HelloJim,
Thanksalotforyourposting.Itisveryhelpful.IhaveanofftopicMinitabquestion,andhopingthatyoucould
helpmeout.DoesminitabhasafunctionlikeVlookupinexcel?
Thanksalot.
Yan
Reply Share
JimFrostAtMinitab
Mod >wuyr 3monthsago
Hi,thankyouforthenicecomment!
Unfortunately,Minitabdoesn'thaveanexactlyequivalentfunction.However,inMinitab,youcanuse
ControlFtousetheFindinDataWindowfunction.Thiswillsearchwithinacolumnforaspecific
value,eitherexactmatchornot.Whenitfindsamatchinacell,youcanlookattheassociated
informationinthethatrowasawaytomimicthefunctionalityofVLOOKUP.
Jim
Reply Share
JackWotton 3monthsago
Hijim,
I'mabletoexplainmyresultsthroughthepvalue,s=,rsq,andthegraphs.butiamunsureonothervalues
thathaveshownupe.g.,DF,SS,MF,F,(howtointerprettheresidualerrortomyresults?whatdoesDF20,
SS235.57MS11.78allmean)ithinkthismostlyrelatestotheanalysisofvarience.hopeyourabletohelp
asihaveadissertationhandinnextmonth)cheers
Jack
Reply Share
JimFrostAtMinitab
Mod >JackWotton 3monthsago
HiJack,
Alotofthesestatisticsarethe"behindthescenes"typeofnumbersthatMinitabneedstocalculatein
ordertocomputethemorecommonstatisticsthatpeopleneed,likethepvalues,Rsquared,
adjustedRsquared,andS.Unlessyouhaveaspecialneed,youoftendon'tneedthestatisticsthat
youlist.
I'llrunthroughthemingeneralforyou.Ifyouneedmoredetailedinformationabouthowthey're
calculated,youcanalwayslookattheMethodsandFormulaHelpinMinitab:Help>Methodsand
Formulas.TheMinitabGlossary(Help>Glossary)alsohasdefinitionsoftheseterms.
DF:Thedegreesoffreedom(DF)describetheamountofinformationyourdataprovidethatyoucan
"spend"toestimatethevaluesofunknownpopulationparameters,andcalculatethevariabilityof
theseestimates.Degreesoffreedomareaffectedbythesamplesizeandthenumberofparameters
inyourmodel.Increasingyoursamplesizeprovidesmoreinformationaboutthepopulation,and
13/22
5/18/2015
inyourmodel.Increasingyoursamplesizeprovidesmoreinformationaboutthepopulation,and
consequentlyincreasesthedegreesoffreedompresentinyourdata.Addingparameterstoyour
model(byincreasingthenumberoftermsinaregressionequation,forexample)"spends"
informationfromyourdata,andlowersthedegreesoffreedomavailabletoestimatethevariabilityof
theparameterestimates.
seemore
Reply Share
JackWotton>JimFrostAtMinitab 3monthsago
Thankyousomuchforyourhelp:)
Reply Share
Fardeen 3monthsago
HiMrJim.
Imhavinggreatproblemsindoingmydissertation.Idontknowhowtomakeuseofregression.Iwouldbe
gratefulifyoucouldhelpme.
Isthereasitewhereitshowsclearlytouseregression?
Thanks
Reply Share
JimFrostAtMinitab
Mod >Fardeen 3monthsago
HiFardeen,
Irecommendthatyoureadmyregressiontutorialwithexamples.Ithinkthiswillansweralotofyour
questions.
Bestofluckwithyourdissertation!
Jim
Reply Share
becbec>JimFrostAtMinitab 2monthsago
HiJim,thankyousomuchfortheinformativediscussionshere.Iammakingmythesis
however,Iamfindingdifficultiesininterpretingmydata.Whatdoesthisresultmeanifmy
constanttvalueis7.114,pvalue=.000,LIFCAStvalue=10.228pvalue.000,LERIANSt
valueis2.971pvalue.003,andEFCOStvalue,2.186andpvalue.029.iwouldappreciate
yourhelp.thanks.
Reply Share
JimFrostAtMinitab
Mod >becbec 2monthsago
Hi,
Withtheinformationyouprovide,Ican'tbesurethatyourmodelmakessense
theoreticallyorwhetherthemodelprovidesanadequate,unbiasedfittothedata.One
thingyoushoulddoisdefintelycheckyourresidualplots.
Assumingthemodelisgood,here'swhatyou'vegot.
Youhaveaconstanttermthatissignificantlydifferentfromzero.However,the
constanttermusuallyhasnomeaningfulinterpretation.There'salinktoablogpostI
wroteaboutwhythisistrueneartheendofthisblogpost(beforethecomments
section).
14/22
5/18/2015
Youhave3significantpredictors.Thissuggeststhatchangesineachpredictorare
relatedtochangesintheresponse.Forexample,aoneunitincreaseinLIFCASis
relatedtoanincreaseinthemeanresponsevalueequaltotheLIFCAScoefficient.
SameforLERIANS.ForEFCOS,everyoneunitincreaseisrelatedtoadecreasein
themeanresponse(youdidn'tincludethecoefficientsbutfromthetvalueIknowthat
theEFCOScoefficientisnegative).
Typically,youdon'tneedtoworryaboutthetvaluesandinsteadfocusonthep
valuesandcoefficients.
Youmightwanttoreadmyblogpostaboutchoosingthebestregressionmodelto
helpyoubesurethatyoudohavethebestmodel!
Bestofluckwithyourthesis!
Jim
Reply Share
dunmao>JimFrostAtMinitab 2monthsago
HiJim,
Couldyoupleasegivemeadirectionforthefollowingquestion?
Myquestions:Iamdoingridershipmodelingusingmultiple
linearregressionmethodinExcelsoftware.Mydependentvariableisboardings,
threeindependentvariablesarepopulation,feederbusservices,andemployment
data.Eventhoughtheconstantismeaninglessdiscussedfromyourdiscuss
group.Inmycase,thepvalueforYinterceptis0.6(greatthan5%),however
theYinterceptcanminimizetheresidual(observeddatapredictedvalue).
Seetheregressionresult:
RSquare=0.943573,
PvalueforYintercept=0.6,Pvaluesforthethreeindependentvariablesare
lessthan5%
AccuracyValidationwithoutYintercept(ObservedPredicted):
seemore
Reply Share
JimFrostAtMinitab
Mod >dunmao 2monthsago
Hi,Irepliedtoyourquestionintheotherpostwhereyousharedyourcomment.You
canfindithere.
Theshortansweris,yes,youshouldalmostalwaysincludetheconstantregardless
ofthepvalue!
Jim
Reply Share
HiJim,
15/22
5/18/2015
HiJim,
Thankyousomuchforyourquickresponse!
Iwanttoincludetheconstanteventhoughthepvalueoftheconstantisgreatthan
5%.Theconstantcanbeexplainedasanadjustedfactorinmypredictionmodelto
minimizetheerror.
Youranswerconfirmsmytestresults.
Thanksagain,
Hope
Reply Share
JimFrostAtMinitab
Hi,you'reverywelcome!
Justtoclarifyonepoint.Yougenerallyshouldincludetheconstantregardlessofthe
pvalue.Youdon'tneedajustificationtoincludetheconstant.Instead,youneeda
verystrongjustificationtoevenconsidernotincludingtheconstant.
Infact,I'veneverpersonallyworkedwitharegressionmodelwhereIfeltjustifiedto
notincludetheconstant.Aregressionmodelwithouttheconstantisveryrare
becausethepotentialforintroducingbiasisveryhigh.
Jim
Reply Share
HiJim,
Icomeback.IhaveanotherpredictionmodelwithYinterceptpositive.Seethe
followings:
AccuracyValidationwithoutYintercept(ObservedPredicted):
Predictedmodel:
DV_37pm=0.441*IV2+0.179*IV3
Error=Observed(3559)predicted(3961)=402(overestimated1678)
AccuracyValidationwithYintercept
Predictedmodel:DV_37pm=0.441*IV2+0.179*IV3+0.714
Error=Observed(3559)predicted(3971)=412(overestimated412)
seemore
Reply Share
JimFrostAtMinitab
Hi,
16/22
5/18/2015
Hi,
YoushouldalmostalwaysincludetheYinterceptinthemodel.Irecommendthatyou
doleaveitinthemodel.
ThisistrueregardlessofthePvalue.Ihopeyou'vereadmypostaboutthe
regressionconstant?Ishowthereasonswhyyoushouldalwaysincludeitinthe
model.
Ifyou'reboundanddeterminedtoconsiderremovingit,thereareimportant
considerationsyoumustevaluatefirst.
1)Checkthestandarderroroftheregression.TheErrorisinyouroutputisnotthe
standarderrorbecauseSisalwayspositive.Yourerrorreductionisnotsubstantial
anywayonlyfrom412to402.Theminisculereductioninerrorsuggestsyoumight
aswellleavetheconstantinthemodel.
2)Checkyourresidualplots.Inparticular,besurethattherearenononrandom
patternsforeithermodel.Thisisespeciallyimportantinthemodelwithouttheconstant
becauseoftenremovingtheconstantintroducesabiasthatyou'llseeintheresidual
plots.Ifyouremovetheconstantandyouseeapatterninresiduals,puttheconstant
backinyourmodel.
But,really,youshouldincludetheconstantevenwiththehighpvalue.It'snothurting
anythinganditislikelyhelpingreducebiasinyourmodel.
JIm
Reply Share
ThankyouJim!
Iwanttolearnmore,soIcomparethetwocasesWiththeconstantinmyprediction
modelANDwithouttheconstantinmypredictionmodel.
WITHtheconstantinmypredictionmodel:
Standarderror:82
Residualplot:73.08%ofprobabilityoutputofthesampledatafitsanormaldistribution.
WITHOUTtheconstantinmypredictionmodel:
Standarderror:78
Residualplot:73.08%ofprobabilityoutputofthesampledatafitsanormaldistribution.
Therearenononrandompatternsforeithermodel.
Frommyunderstanding,theconstantissmall,sothereisnopatternintheresiduals
distributions.
Lastquestion:iftheconstantisbig,itcausestheerrorreductionsubstantial,doIstill
needtokeeptheconstant?(sorry,Idon'thavetheregressionresults,butIwantto
knowifthecaseexists.)
Thankyou,
Hope
17/22
5/18/2015
Hope
Reply Share
JimFrostAtMinitab
Mod >dunmao amonthago
Hi,
Givenwhatyousay,theredoesn'tseemtobeanynumericreasontonotremovethe
constant.
However,beforeyoudothat,askyourselfifit'stheoreticallyjustifiedthatifyousetall
ofthepredictorstozero,you'dexpecttheresponsetoequalzeroaswell.Preferably,
youwouldalsohavemeasuredvaluesnear/atthisallzeroregiontoconfirmthatthe
regressionlinetrulygoesthroughtheorigin.
It'sonlywhentheconstantissmallthatyouhaveachance(smallchance)toremove
itfromthemodel.Ifitislarge,removingitfromthemodelwillalmostcertainlybias
yourmodel!Iwouldneverremovealargeconstant.
Jim
Reply Share
dunmao>JimFrostAtMinitab amonthago
HiJim,
Thankyousomuchforyourexplanation!Icompletely
understandtheconstant(regardlessofpvalue)now.
NowIhaveanewregressionresult:
R=99.35%,
AdjustR=99.06%
DV=20+0.129*IV1+0.178*IV2+0.078*IV3
Errors=observed(4088)predicted(4052)=36
Averageerrors=5.75%
Questions:WhyistheRsobigat99.35%?Maybesomeonewouldaskmeabout
thequestion.However,thisistrueregression
result.Howwouldyouexplaintheresult?
Thankyouagain,
Hope
Reply Share
JimFrostAtMinitab
Mod >dunmao amonthago
Hi,
Withoutknowingthespecificsofthemodelandthestudyarea,it'simpossibletosay
forsure.IfIremembercorrectly,youaremodelingridershipovertime.Ifthereare
trendsinthedatathataffectbothsidesoftheequation,thisisaproblemandcanoften
produceinflatedRsquaredvalueslikethis.Youshouldplotthevariablestoseeif
18/22
5/18/2015
produceinflatedRsquaredvalueslikethis.Youshouldplotthevariablestoseeif
theyarestationary(constantmeanandvarianceovertime)ornonstationary(upward
ordownwardtrendornonconstancevariance).
Ifyouhavenonstationarydata,youmustmakeitstationarybydifferencingthedata
sothateachdatapointisthechangeinvaluebetweenconsecutivepoints.Using
regressionanalysiswithtimeseriesdatainvolvesadditionalconsiderationslikethis.
Unfortunately,Idon'thaveahandyreferencetoreferyoutoobutyoushouldperform
someadditionalresearchtoensurethatyouendupwithavalidmodel.
Jim
Reply Share
dunmao>JimFrostAtMinitab amonthago
HiJim,
Inoticedanewquestion:
AsItoldyouIhavedonethetestingasthefollows:
=====================================================
Whenletintercept=0,theregressionresult:
Rsquared=0.96
AdjustedRsquared=0.88
StandardError=78
Observations=14
ANOVA:
df
Regression:2
Residual:12
seemore
Reply Share
Em 2monthsago
Hi,thankyouforyourextremelyhelpfulblogs!Iwaswondering,ifyoucanhelpmeoutwithmymultiple
regressionanalysis.ForthePearsoncorrelation,Ifoundthatonlyoneofmypredictorsissignificant
(p=0.037).However,Idon'tquiteunderstandwhyinthettestsection,noneofmyindependentvariables
makeasignificantcontributiontothemodel.Howisitpossible?Icouldn'tfigureoutthelinkbetweenthetwo.
Canyouexplainthis?Thanksinadvance!
Reply Share
JimFrostAtMinitab
Mod >Em 2monthsago
Hi,
ThePearsoncorrelationpvaluesandregressionpvaluestestdifferentthingssotheanswersmay
notagree.Thecorrelationpvalueonlytestsonepairofvariablesatatimewithoutconsideringthe
othervariables.Theregressionpvaluesfactorinalltheotherpredictorvariablesthatareincludedin
themodel.
Fromwhatyouwrite,itsoundsasthoughthecorrelationpairthatisissignificantisoneofthe
19/22
5/18/2015
Fromwhatyouwrite,itsoundsasthoughthecorrelationpairthatisissignificantisoneofthe
predictorsandtheresponsevariable.Tryaregressionmodelwithjustthatonepredictor.Itshouldbe
significantinaregressionmodelbyitself.Then,addintheotherpredictors.Ifthesignificancegoes
away,itindicatesthattheotherpredictor(s)areaccountingforsomeofthesamevarianceinthe
response.Bysplittingupthevariancethatisaccountedforbetweenthevariables,itmaybethat
nonearesignificantwhenthereismorethanoneinthemodel.
Also,checkyourVIFsinthefullmodel.It'spossiblethatmulicollinearity(correlationbetweenthe
predictors)issappingthesignificanceofthepredictors.Theproblemsassociatedwith
multicollinearitydonotoccuronlywhenthereisastrongcorrelationbetweenindividualpairsof
predictors.Theseproblemscanoccurwhenthereisamoderatecorrelationbetweenanumberof
predictors.ThismoderatecorrelationmaynotbesignificantwhenyoulookatthePearsoncorrelation
betweenpairsbutcanbedetectedwithVIFs.Readmoreaboutthisinmypostaboutmulticollinearity
andVIFs!
Ihopethishelpsandthanksforwriting!
Jim
Reply Share
Sayeed 2monthsago
HeyJim,
howdoyouinterpretanadjustedRSquareresult.Foreg,Ihadtofindthecorelationbetweenexchange
rateandstockprice,ItgavemeananswersayingtheadjustedRSquaretobe0.3925.Isthereacorelation
andifthereisthanhowdoyouwritethat?
Thanksinadvance
Reply Share
JimFrostAtMinitab
Mod >Sayeed 2monthsago
HiSayeed,
That'sagreatquestion!
I'vewrittenabouthowweoftenuseadjustedRsquaredtohelpincludethecorrectnumberof
predictorsinthemodel.
However,thereisaspecificinterpretationforadjustedRsquare.AdjustedRsquaredprovidesan
unbiasedestimatedofthestrengthoftherelationshipbetweenthepredictorsandresponse.
RegularRsquaredisthestrengthofrelationshipinyoursamplebutitisabiasedestimateofthe
populationbecauseittendstobetoohigh.AdjustedRsquaredis"shrunken"soitisnotbiased.
Foryourresults,themodelaccountsforanestimated39.25%ofthevariabilityintheresponseinthe
population.WhatevervaluetheregularRsquaredis,itonlyappliestoyoursample.
IwroteanentirepostaboutthisthatIrecommendyouread:Rsquaredshrinkage.
Thanksforwriting!
Jim
Reply Share
Javaid>JimFrostAtMinitab 2monthsago
Ihaveaquestion:
20/22
5/18/2015
RegressionEquation
MR=0.00349+0.003154A+0.16467B+0.000595C
andthatgivesmeModelSummary
SRsqRsq(adj)Rsq(pred)
0.001568899.56%99.30%98.23%
AmIcorrectinassumingthatthevalueofRsqis0.9956?
Who We Are
Authors
Minitab is the leading provider of software and services for quality

improvement and statistics education. More than 90% of Fortune 100
companies use Minitab Statistical Software, our flagship product, and
more students worldwide have used Minitab to learn statistics than any
other package.
Carly Barry
http://blog.minitab.com/blog/real
worldquality
improvement
Patrick Runkel
http://blog.minitab.com/blog/statistics
andqualitydata
analysis
Joel Smith
http://blog.minitab.com/blog/fun
withstatistics
Kevin Rudy
http://blog.minitab.com/blog/the
statisticsgame
Jim Frost
http://blog.minitab.com/blog/adventures
instatistics
Greg Fox
http://blog.minitab.com/blog/data
analysisand
quality
improvementand
stuff
Eric Heckman
http://blog.minitab.com/blog/starting
outwithstatistical
software
Dawn Keller
http://blog.minitab.com/blog/adventures
insoftware
development
Eston Martz
http://blog.minitab.com/blog/understand
Minitab Inc. is a privately owned company headquartered in State College,

Pennsylvania, with subsidiaries in the United Kingdom, France, and
Australia. Our global network of representatives serves more than 40
countries around the world.
Visit Us at Minitab.com
Blog Map http://blog.minitab.com/sitemap.html | Legal
http://www.minitab.com/legal/ | Privacy Policy
http://www.minitab.com/legal/#privacypolicy | Trademarks
http://www.minitab.com/legal/trademarks/
Copyright 2015 Minitab Inc. All rights Reserved.
21/22
5/18/2015
statistics
Karen Meldrum
tipsfroma
technicaltrainer
Bruno Scibilia
http://blog.minitab.com/blog/applying
statisticsinquality
projects
Eduardo Santiago
http://blog.minitab.com/blog/understand
statisticsandits
application
Cody Steele
andquality
improvement
22/22

How To Interpret Regression Analysis Results - P-Values and Coefficients - Minitab PDF

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

How To Interpret Regression Analysis Results - P-Values and Coefficients - Minitab PDF

Transféré par

Droits d'auteur :

Formats disponibles

5/18/2015

The Minitab Blog

Data Analysis http://blog.minitab.com/blog/dataanalysis2

Regression analysis generates an equation to describe the statistical relationship between

How Do I Interpret the PValues in Linear Regression

How Do I Interpret the Regression Coefficients for Linear

How Do I Interpret the Regression Coefficients for

If you're learning about regression, read my regression tutorial

You Might Also Like:

Name: Henry Mwangi Thursday, February 20, 2014

Name: Deeps Dee Thursday, March 27, 2014

Name: taiwo lucas Wednesday, April 2, 2014

Name: O.Jobi Saturday, May 10, 2014

Name: yashika Tuesday, May 13, 2014

Name: omid Saturday, June 7, 2014

Name: Jim Frost Monday, June 9, 2014

Name: Mrv Yrd Wednesday, August 27, 2014

Name: Jim Frost Thursday, August 28, 2014

Thanks for reading!

Mod >inez 6monthsago

Mod >Cain 5monthsago

Mod >WDC123 4monthsago

Mod >Ronja 4monthsago

Mod >Fiachra 4monthsago

Mod >Fiachra 4monthsago

Mod >sewnsew 4monthsago

Mod >Scott 4monthsago

Mod >PatrickKajubili 4monthsago

Mod >Marija 3monthsago

Mod >wuyr 3monthsago

Mod >JackWotton 3monthsago

Mod >Fardeen 3monthsago

Mod >becbec 2monthsago

Mod >dunmao 2monthsago

Mod >dunmao 2monthsago

Mod >dunmao 2monthsago

Mod >dunmao amonthago

Mod >dunmao amonthago

Mod >Em 2monthsago

Mod >Sayeed 2monthsago

Minitab is the leading provider of software and services for quality

Minitab Inc. is a privately owned company headquartered in State College,

Vous aimerez peut-être aussi