Let's talk about stats baby!

"Let's talk about stats baby!

Let's talk about what Mu means!.....

Let's talk about stats!"

So, yeah, the opening remarks really do sum up this post. I can't begin to stress how important understanding, (I mean really understanding), statistics is for data science! It doesn't take much to run a regression and get the output in R, Excel or any other software, but interpreting the results clearly can take you far. (that rhymed!)

Learning statistics isn't about memorizing formulae, and though it uses a lot of greek letters it isn't all greek! I'd like to call statistics "Applied common sense" because in the end, beyond the T-tests and P-values, that's what it is.

So, if you're an absolutely total beginner and you've never even heard of regression this is the book for you: Business Statistics: Communicating with Numbers

It's a very basic book, explains the concepts pretty well and includes a section on statistical writing (which is extremely useful). when I mean the basics, they start from the very definition of statistics and build all the way up to logit and probit models.

I do think the ANOVA (Analysis of Variance) chapter, could've been slightly better explained. I will add resources to learn specific topics in the future.

Pro tip: Whenever you encounter fancy equations, ask yourself why we do this or use these equations. What is it telling you?

Now, if you are already a little confident with your basic probability, and regression techniques, you might want to try and have a go at this: John E. Freund's Mathematical Statistics with Applications

It's a wonderful and much mathier book than the previous one, but if you are up for the challenge, go for it!

A popular approach to machine learning is statistical learning. It's basically applied statistics and the techniques are what I like to call the gateway into Machine Learning and Data Science for the economist. Biostatistics is the equivalent for biological sciences. In this post, I just have an introductory econometrics book to suggest.

Basic Econometrics by Damodar Gujarati is as the title suggests, basic.
I haven't really found any other books that  I've felt are super comprehensive for the first time learner of econometrics. As, when learning econometrics, it's good to have a little knowledge of micro, macroeconomics and basic stats as well.

If you are just looking for a quick refresher The Little Handbook of Statistics might be what you are looking for. There are tons of video courses out there on Coursera, Udemy and MIT OCW, and Khan Academy where you can choose the course as per your needs.

For beginners, I recommend this course on MIT OCW. It is a little challenging if you are stepping out into this world for the first time, but it is a challenge where you are highly rewarded.

Now, this next resource might sound a like a bit much for a beginner, but I find that it has THE best explanations for basic concepts of statistics such as normal distributions, p-values and regression. It is hands-on with an R session too.
Introduction to Statistical Learning with R  is the best resource for machine learning and offers the crispest and clearest concepts of basic regression. It is a beautiful book and also part of a free online course taught the authors themselves offered by Stanford.
You can always save this for later if you feel like it's too much as they dive into examples of KNN and Logistic regression pretty early on. Perhaps, after getting a hang of regression you could try out the R labs in the online course to help reinforce your learning.


That's all I have for you in this post if any of the above resources were too difficult to comprehend at your level please reach out, I'm happy to help!

Happy Learning!