Statistics 101: Nonlinear Regression, The Piecewise Model

In this video, we dive deeper into nonlinear regression and explore piecewise regression. What if we can model different parts of the data using more than one linear model? We learn how to do that and then compare it to the quadratic model using the same data.
My playlist table of contents, Video Companion Guide PDF documents, and file downloads can be found on my website: www.bcfoltz.com
Happy learning!
#statistics #regression #machinelearning

[@user-qu7sh1kb1e]

From the start of PL1 to the end of PL18. Thank you very much for your contribution. It's your encouragement at the beginning of each video that inspire me not give up. The explanation is very detailed and convincing. The graph and interpretation are vivid and insightful. Thank you again. You make my life better and easier.

[@talhaanwer8559]

I strated STATS 101 from PL1 on 26 Jan 2019 and end PL 18 on 5 Sep 2019 ALHAMDOLILLAH. Its quite a long journey for me but really i enjoyed a lot during a whole peroid. Thank you very much for your wonderfull series of lecture. It's your encouragement at the beginning of each video that hold me not to give up. The explanation is very detailed and convincing. The graph and interpretation are very meaningful and insightful. Now this helps me alot for my DATA SCIENCE career. Thank you once again.

[@varunpatwardhan1780]

Thank you so much for making these statistics videos. I was freaking out during my first semester of Masters in Data Science! I had zero stats knowledge. Your videos helped me 1000X more than my classes during my Master's!

[@PriyankaSharma-yh6sb]

I found your channel, now feeling blessed ! waiting for next video.... you are just awesome !

[@samyong307]

Very good explanation for beginners to understand, thank you very much!

[@stan0702]

Thank u for this video series Brandon...it's very enlightening..

[@oanaandreea7525]

I cannot thank you enough! you have no idea how much you helped me!

[@MrLennon70]

so far, this is the best explanation about piecewise regression on KZbin. You deserve more likes and subscribe

[@shortly2103]

Really amazing tutorials.... Statistics made easy and simple

[@ylazerson]

Fantastic video - thanks

[@rhishikeshjoshi6806]

Hey Brandon, Kudos on the video. It's amazing how simple you make these things sound. Simply love it. I just have one question. If I want to know more about the iterative process of finding a knot location how do I do that without just eyeballing? It would be great if you can suggest any content for that. Thanks

[@nasgaroth1]

You're the best ;-) Thanks and good luck

[@ibrahimalbusaidi7284]

You are really excellent in simplifying and explaining all these terms! Can you please make playlist about Structural Equations Modelling (SEM)?

[@lmanuelsantiago]

This are great simple and tp the point thanks!

[@lmanuelsantiago]

I sent you an massage on your web page. Let me know how is best to contact you.

[@lawrencewong9369]

thanks for the explanation detailly. there has one question about whether a quadratic or simple linear regression model is necessary before we calculate the piecewise regression.

[@adeepak7]

Hi Brandon! Can you kindly explain how you calculate all the measures, i.e., the summary output , regression statistics, ANOVA and all other things. Can you kindly make a video on that? It will be a great help.

[@mehmetkaya4330]

Another great tutorial video. T-h-a-n-k Y-o-u !!!!!!!!

[@nilmonysaha3113]

You are the best!

[@anuvaagarwal3492]

Thank you for this wonderful explanation, Brandon! Is there a video on piece-wise regression with multiple knots too?

[@josephcrosby8298]

Hi Brandon, excellent video, do you think you will explore smoothing and general additive modelling at all in future videos?

[@stenlons5317]

Hi Brandon! Is there a way to get the slides of your statistics series? And thank you so much for what you do! Stats are so clear when you explain it!

[@francodauro5081]

Hi Brandon. Great video! Can you tell me about the books or paper that you read to create this lesson. Thank you from Argentina

[@quosswimblik4489]

If you approach the next line equation and ascend from the current line equation as you get closer to the point of intersection you can curve the model where it applies. You can extend this technique into 3D and use the method to turn triangular mesh's into curved surfaces. My problem is I don't have a name for this curving function I'm open to suggestions.

[@enverterrazas5072]

Hi, so what must I do against overfitting? Thanks for your time!

[@medicinalvl-up862]

probably you wont read this but "muchas gracias desde PERU!!"

[@edwinvivek9887]

Do some videos on Regularization regression techniques

[@BrandonFoltz]

L1 and L2?

[@edwinvivek9887]

Yes

[@ChocoLoco47]

At the end of the video you compared the two models based on how much variation they explained. What statistical test would you use to compare these two regression models to check if one indeed explains significantly more variation than the other?

[@antoniaangelopoulou4269]

Many thanks! Could you please interpret the coefficient of Job Tenure^2?

[@theresemeyer9927]

hi! can you please make a video on generalized linear models?

[@ccuuttww]

I don' think it have general method for linear or nonlinear regression if u want maths apporach u can study least square approximation

[@ccuuttww]

I watch your channel since 1-2 year ago my learning schedule is much far beyond this topics however I don't know how to operate it in excel I perfer linear algebra to slove those question but excel seems don;t have this option

[@giDDRu]

how do you calculate the b2 coefficient? and do you actually calculate knot value, or you just put the value you think is good enough? i can't find the instructions on minitab

[@panagiotisgoulas8539]

Which formula would give me the b1 and b2 in this type of regression equation (besides dumping everything in the statistical software). I would have to use the least squares errors method and start all over for that?

[@myronh2066]

I have a question, can we compute the turning knot instead of eyeballing it?

[@heronpereira5754]

Hi, here is a doubt: I'd like to know if is statistic correct to use means as samples. Example: take a bunch of samples during a period of time, make a mean out of it and consider this mean as a sample. Them make statistical analisys base on these samples. Is also any material about this topic? Thanks in advance.

[@user-ff8wj9vr8m]

multiple knots please!

[@nasgaroth1]

Hi man, please make a video about Markov chain

[@BrandonFoltz]

I have several! Just do a search on my KZbin page :)

[@bhavnaphogaat2866]

Can we apply segmented regression on panel data?

[@MrLennon70]

what if i want to make piecewise regression with multiple knots? Can i do that? How?

[@user-fh3gr7sq1z]

nice. is there a code for this model in R ?

[@marcusjohnson5486]

Why two different multiple R values? Isn’t multiple R just correlation?

[@svarthal8845]

Hi sir, we need your help so much. We've a problem in accounting and we cannot solve it. Can you help us, please?

[@svarthal8845]

We've data, and we have to prepare its balance sheet in excel. Can u show us thw ways of transactions?😔

[@lalitajoshi5350]

Sir, in which of your playlists can I find videos on poison arrival and service rates?