Statistics with R (1) - Linear regression

[@badmusidowu379]

There are many great researchers and their works on line. Their researches have been helpful. kudos to everyone.

[@horizon2reach561]

You cannot even believe how great your tutorial is, fantastic and thanks a million!

[@michaelconnors7668]

Thank you. I had no success learning R and have tried for months. I stumbled on to your video and I learn to run names after I load a data set. How simple. Now I am off and running.

[@calledun2glory]

I am transitioning from Stata to R and this video has been very helpful. Thank you

[@godfreyphilliamapangu3129]

Thank you so much Christoph. You have addressed a real problem for me in R. I will continue to follow your videos for more insights

[@vanialed3882]

THANK YOU!!!!!!!!!!!!!!!!!!!! This is incredibly helpful!!! I was ready to fail statistics and honestly this has given me hope and I can save the semester! To me statistics is just sooo difficult to understand, but this was of great help! Thank you! :)

[@ChristophScherber]

+Vania Led you´re welcome!

[@madanlalgupta3026]

Vania Led [.

[@Loachie90]

Vania Led how’d that semester turn out?

[@JoeDoesVoices]

Brilliant, thanks! I’m switching to R from a few years of heavy point-and-click SPSS use and had been feeling slightly intimidated, but your videos are helping me right along.

[@rjfabreu]

Simple! Very good for people like me (1º stepts on R). Thank you Christoph!

[@ProfNandishPatelPhD]

Thank you Christoph! Excellent introduction to R. Simiplicity is most effective for learning. :)

[@spikyboxable]

The most easy to follow but enough to understand the model. Thank you so much.

[@sn9751]

Thank you very much for this explanation, You are brilliant and very good mentor

[@Albo-bot]

An excellent lesson thank you Christoph

[@Liurika]

This is great!! Now I have a much better idea of what I'm doing in R. Thank you very much!!

[@ChristophScherber]

Thanks also! It´s great to see so many people becoming interested in R.

[@percydai8131]

Very easy to follow instructional video!

[@msheen99]

The lesson was very simple and informative!.Thanks a lot Chris

[@padula77]

Tks Sir! Very well explained. I have a Bsc in statistics but I need to get hands on R.

[@29160bubu]

Many thanks Mr Scherber.

[@zezeandjr4110]

Very insightful, smart yet easy to follow, I can immediately see how I can use that in my line of work, I just need to broaden my knowledge of the language first. Many Thanks

[@sahadulislam2380]

Bangladesh moduli

[@carolinekundu8215]

Wow, thank a million tonnes, been trying to split my head on this R analysis

[@silbercharles]

Very very helpful in explaining linear regression. Thank you.

[@hanishadesu3165]

you are really awesome.... so clearly explained everything.......thankyou so much

[@norhasyimahseliman6856]

TQ so much. Easy to understand.

[@shalinibhawal781]

Thank you!! Found it very helpful.. Being able to pause and try commands helped my learning..

[@rajeevsuresh9542]

Awesome job man! This is good stuff, thank you for this!!

[@KnorpelDelux]

Super Video, vielen Dank aus Bayern!

[@ChristophScherber]

gerne!

[@dominiquebarrette9621]

Truly informative, thank you, you explain so well

[@mengshayin2201]

thank you Christoph so much

[@ChristophScherber]

Thanks! Great that you liked it. Let me know what else you´d like to hear!

[@AcrylDame]

This was really a great explanation. Thank you! Subscribed

[@Testqqqqqqq]

Easy and good starting point on R

[@pablofreudenthal]

GREAT video! It's being a great help in my research

[@creamcrackers279]

Thank you so much. I like the way u took ur time to explain things slowly. Thank u once again

[@jibreelahmad1682]

woow! what an incredible tutorial. This the best, I really appreciate your work Christoph :-)

[@ChristophScherber]

many thanks!

[@jill6380]

THANK YOU! You made it so easy! Now, please teach my professor how to teach.

[@ChristophScherber]

thanks! well, I had a great teacher, too - so maybe that´s the trick...? Best wishes, Christoph

[@elliotsalsa3561]

Thanks, A great Teacher

[@Drzhivago11]

Very nice! Good job..

[@ericjohnson1811]

Excellent example. Thank you for posting!

[@charlesaydin2966]

Vielen Dank, sehr sehr nuetzlich!

[@adancastro2220]

great explanation... Thanks

[@sanjayjoshi4531]

too good.searching for a similar video on non parametric methods for testing of hypothesis and also FOR SPRT

[@rebeccaliu2049]

Thanks a lot, this is really easy to follow.

[@eirikur09]

This was awsome, cant wait to see your next video

[@ergosumdre]

Great content.

[@oleksandrsholin329]

Thank you very much, everything works pretty well, now I understand at least some stuff in R.

[@KristinaPetersenKarlsson]

I totally agree with AcrylDame! Thanks!

[@Goddebates]

Best video on this topic

[@SeedGeorge]

great video, thanks

[@carlosbrown4202]

Could you explain how to get the method of moments estimator for the compound poisson process?

[@henriquevonhertwigbittenco6057]

Thank you Christoph... I just had my first R class!

[@francescopaderi3356]

Henrique von Hertwig Bittencourt f.w’w’w’w Ww

[@francescopaderi3356]

P’

[@francescopaderi3356]

Henrique von Hertwig Bittencourt th

[@francescopaderi3356]

,le.s’e’slslls

[@8dholland]

Really good tutorial. I usually don't say this either

[@AmonMwaijengoForester86Mwai]

hello, how do we deal with diversity indices in R??. Thanks.

[@RajvimalCrystal]

perfect for beginners.. i appreciate your work

[@helmutdittrich3420]

Awesome, thank's a lot. Very helpful

[@gbaremz3633]

thank you..you have really been of great help to me

[@johnkeen4426]

Thanks so much for posting this!

[@RyanBenkeser]

Great video. It looks like Stata is very transferable to R.

[@ditke71]

Very nice prezentation, though for me is not clear how p-value is related to the t value or maybe to the level of significance. The null hipotheses is that the parameter (coefficient is equal to zero). So when do we reject this?

[@LinuxBadukNMore]

Great video. I sure hope your thinking of making more.

[@adomderkye4743]

In fact, you have explained things well. Could you, please, illustrate how to input data from excel into r.

[@ChristophScherber]

thanks, I´ll try to cover this in a future video!

[@Jjzein]

nice interpretation with right execution. Thank You

[@flamboyantperson5936]

Hello, please come back and make more vidoes on R we really need you Sir.

[@manicfre]

Hi, Did you make any video on emmeans? I am trying to calculate p-value and making a vertical bar chart but I am unable to do it. Could you please help me out? Thank you.

[@samukelisodlamini798]

After writing data which button u press to pops up data set range plz help me???????

[@roshanfernando9957]

Does anyone have any experience with Version 3.3.0? I am having trouble with the "control R" part and calling data...

[@robins5458]

Hey Chris, thanks for your video's. I am having a problem... I have R3.1.2 I am attempting to follow you vid and its keeps giving me an error. data().. is good names(airquality) is good but when I go to plot plot(Ozone-Solar.R,data=airquality) I get error... object not found. Error in plot(Ozone - Solar.R, data = airquality) : error in evaluating the argument 'x' in selecting a method for function 'plot': Error: object 'Ozone' not found am I just missing something? thanks Robin

[@rybavresu9917]

It seems you are using "-" instead of "~". When I applied your script, the error message was the same. With "~", however, it works perfectly.

[@mmabath]

@Christoph Scherber, thank you for video, it is really helpful. quick question, how do i run a regression plot that has labels for two specific points on the plot. in other words, how do i label specific points on the plot?

[@ChristophScherber]

You can use the text() function in combination with the locator() function for that

[@GrisKatya]

Great lesson ....thank you very much

[@MrBenFranco]

Tks Mr. Scherber!

[@sdgarcia34]

These are all great and very informative videos! Thank you very much for posting them. Still i am straggling a bit, I would like to check the correlation of let say 8000 genes in 10 different patients that I organized from less sick to more sick (let say percentage of neuronal loss). There is any way I can get the correlation of the expression of these genes with the sickness of the patient looking at the r? (ex. 0.99 would be a gene that correlate better with a worse pathology, right?) if so, Can I check all the r for each gene using R?

[@delt19]

Very helpful tutorial. Any tips on sourcing data directly from an excel spreadsheet(s) to run regressions in R?

[@delt19]

Wow what a noob question from a younger me.

[@sicktoaster]

I got the console stuck to the size of the application window. Now when I attempt to load an R file it replaces the console. I can switch back to the console but I can't get it to display them both at the same time. How do I get it back?

[@wongibedieudonne3778]

thanks chris

[@gabriellejohnson5218]

How can I calculate r of the dataset?

[@TSBilliards86]

I can't seem to produce a linear regression graphs with 64bit. I can plot a normal scatterplot, but not the second line.

[@ChristophScherber]

This is usually related to the machine you´re working on; what operating system are you using?

[@TSBilliards86]

Christoph Scherber I have win 8.1 (64 bit)

[@aberaatumo9361]

thank very match. i have got knowledge

[@amaral-obaidi4549]

Thank you very much vielen Dank für Ihr Helfen grüße

[@lotto009]

You number ONEEEEEEEEEEEE Thank you

[@namangupta5840]

hey man why dont you create more videos. you explain very well i am sure that your channel will roll

[@ChristophScherber]

I´ll try as soon as possible!

[@zubinsarkari5994]

please someone tell me what is the sign after the ozone while plotting im not able to plot the graph of Ozone and Solar.R

[@ChristophScherber]

When plotting, you usually use the tilde symbol "~". Cheers, Christoph

[@zubinsarkari5994]

Thank you so much Christoph Scherber. Your explaination regarding the linear model was superb looking forward to see more of your statistical analysis. Cheers!

[@charlesrambo7845]

Thank you!

[@ShubhratSharma]

exceuteing abline(h=mean.Ozone) is giving error Error in int_abline(a = a, b = b, h = h, v = v, untf = untf, ...) : cannot coerce type 'closure' to vector of type 'double' what should i do??

[@vamshiindla3514]

Probably mean.Ozone is not calculated as per the video..please check. I was able to perform as per video.

[@kerli1551]

How can you tell that the relationship is not linear from looking at the plot?

[@danig7253]

Thank you for this video, so useful. Do you have anything for multivariate multiple regression by any chance?

[@ChristophScherber]

+Dani G Thanks, I will put this on my to-do list!

[@danig7253]

Fantastic, thank you Christoph. Do you know of any other tutorials or reading materials relating to multivariate multiple regression that might be useful for someone who is new to R by any chance? I'm scrambling around at the moment trying to make some sense of it for my thesis and I am more used to working with SPSS but from my understanding, a MvMR is not possible on SPSS unfortunately!

[@ChristophScherber]

There are some great introductions by John Fox (McMaster University); however, the problem in these cases is that you need to decide whether you´d like to look at contrasts on the side of the response matrix Y or the explanatory variables X. Let me know if I can help. Cheers!

[@danig7253]

That's great, thanks again Christoph. Due to timelines, I'm thinking I might use canonical correlation (which is possible on SPSS) at this stage (submission of my thesis) and then perhaps try something more sophisticated for later attempts to publish!

[@enian82]

Why do I get this error "Error in lm(Ozone ~ Solar.R, data = airquality) : unused argument (data = airquality)

[@twinklegowani2546]

Christoph I would like to ask how do I my own data set, instead of using the data set provided by R studio ?

[@sunilbobb]

Thanks a lot sir

[@riteshmehrotra9901]

Nice explanation, one thing wanted to confirm though - did this model showed linear relation between ozone and solar radiation, how can this be confimed?

[@ChristophScherber]

Well, you´d usually look at a plot of the residuals vs. the fitted values (the first of the plots appearing when typing "plot(model)"); if you see non-linear patterns in this graph, then you can conclude that your model may contain nonlinear terms.CheersChristoph

[@nouraldinhalayqa7646]

Christoph Scherber How can we get Tilda symbol

[@ChristophScherber]

usually you can find it close to the "plus" sign on your keyboard (depending on your keyboard layout).

[@Bubba83]

Hi. I was wondering about the interpretation of coefficients, from a linear regression model that has iv's that are categorical. In plain language, how would you interpret the output itself? For example, dv = nights in hotel, iv = gender (male/femle), you run a lm(nights~gender) and your output is something along the lines of gendermale coeff = 2. What does this mean? That by being male, your more likely to stay two nights longer in a hotel etc.? Thanks for any unput and nice video.

[@ChristophScherber]

Hi Oliver,Depending on whether you have interaction terms present in your model, the coefficients are either differences in intercepts, and/or differences in slopes.All categorical iv´s are internally re-coded as so-called dummy variables (e.g. zeros and ones). In your example:model1=lm(nights~gender)summary(model1)(Intercept) 1gendermale 2This means: A female spends on average 1 night; a male spends 1+2=3 nights. The "gendermale" coefficient is therefore the *difference* between males and females, or the "extra-number-of-nights" spent by a male compared with a female.The same principle applies to more complex models!Cheers,Christoph

[@kuldeepkurroliya2555]

Sir, Please post the tutorial for user defined function for Simple Linear Regression in R without using any R packages or library.

[@niranjanpatil7618]

nice video could you please tell me how to predict label using 3 attributes and one attribute is qualitative

[@EndiusJN]

Thank you so much!

[@SagarmathaSipahi]

I am sure this is a great tutorial but in my case when I use the plot command, it keeps on saying object 'Ozone' not found (Please help, if possible): My editor: data(airquality) names(airquality) #Ozone" "Solar.R" "Wind" "Temp" "Month" "Day" plot(Ozone,Solar.R, data=airquality) My R Console > data(airquality) > names(airquality) [1] "Ozone" "Solar.R" "Wind" "Temp" "Month" [6] "Day" > plot(Ozone,Solar.R) Error in plot(Ozone, Solar.R) : object 'Ozone' not found > plot(Ozone,Solar.R, data=airquality) Error in plot(Ozone, Solar.R, data = airquality) : object 'Ozone' not found >

[@ChristophScherber]

you need to type plot(Ozone~Solar.R,data=airquality). That is, the tilda "~" was missing

[@junruizhang9454]

This video is extremely helpful! Thanks a lot! Can I have anther question for you? Assume I have 3 columns. Id,price and date. I want to run regression model between price and date, but under the condition that they have the same id. In another word, I want to see all the regression result that are grouped by different id. What codes should I put there? Wish you could help me! Thanks a lot!

[@ChristophScherber]

this looks like a mixed-effects (hierarchical) model could do the job!

[@robins5458]

I got the answer from stack overflow.(Gregor de Cillia) the - is really a ~ thanks

[@jwalexandre11]

If using a newer version, ~ works but - is not working

[@edwinkairu9882]

Yeah, I noticed that.

[@amirdavidovich2425]

awesome! ty

[@guramritaujla6930]

Thanks a ton, ur awesome

[@abbas8646]

Thanks very much indeed

[@radiazionecosmicadifondo1122]

SUPERefficient!!

[@AliJwan]

thank you