Questions on my online forecasting course

I’ve been get­ting emails ask­ing ques­tions about my upcom­ing course on Fore­cast­ing using R. Here are some answers.

Do I need to use the Rev­o­lu­tion Enter­prise ver­sion of R, or can I use open-​​source R?

Open source R is fine. Rev­o­lu­tion Ana­lyt­ics is orga­niz­ing the course, but there is no require­ment to use their soft­ware. I will be using open source R with Rstu­dio for demon­strat­ing things in lectures.

Is the course free?

No. It costs $1000 to reg­is­ter for the course. The course is being run jointly by Rev­o­lu­tion Ana­lyt­ics and Monash Uni­ver­sity. Most of the money will go to Monash Uni­ver­sity, and will fund a research assis­tant to help me with R pack­age devel­op­ment. My R pack­ages and my FPP text­book are free, but they do take time and resources to pro­duce. This course is one way I am fund­ing them.

Can I watch the lec­tures in my own time?

Pro­vided you are reg­is­tered, you will be able to log in and watch the lec­tures after the event. If you really want to, you can watch them repeat­edly. This is par­tic­u­larly impor­tant for peo­ple in East­ern Europe or West Asia where the lec­tures are in the mid­dle of the night.

Will I be able to ask questions?

Yes. The lec­tures are meant to be inter­ac­tive, for those watch­ing live. You can ask ques­tions, inter­ject, etc. I will also try to respond to ques­tions after lec­tures, espe­cially for those who can­not watch live.

If I’ve already read your books and blog, will I learn anything?

Prob­a­bly. The course is based on my FPP book, so don’t expect me to cover things that aren’t dis­cussed there. But most peo­ple find that they learn more through ask­ing ques­tions, see­ing exam­ples worked out, dis­cus­sion, etc.

I have very lit­tle expe­ri­ence with R, short of play­ing around with Rcmdr. Is that ok?

It is worth brush­ing up on R if you don’t use it reg­u­larly. I will not be using Rcmdr. Instead, I will use Rstu­dio and we will be learn­ing the rel­e­vant com­mands. A suit­able intro is the tuto­r­ial at www​.otexts​.org/​f​p​p​/​u​s​ing-r.

If you feel that is not enough, try work­ing through the tuto­ri­als at cyclismo​.org.

Is there a read­ing list I can start on lead­ing up to the lectures?

The course is not dif­fi­cult from a mathematical/​statistical per­spec­tive, and as long as you are not freaked out by a few equa­tions you should be fine. If you wanted to brush up, go over mul­ti­ple regres­sion at an intro­duc­tory level. A suit­able back­ground on this is chap­ter 5 of FPP, or chap­ter 3 of Pardoe’s Applied regres­sion mod­el­ing. But you may have some­thing else handy that would be equally suitable.

What else can I do to prepare?

If you really want to get a head start, try work­ing through Michael Lundholm’s tuto­r­ial on R’s time series facil­i­ties.

I want to learn about fore­cast­ing xxxx. Is this course use­ful for this purpose?

The course is about time series fore­cast­ing. That is, when you want to fore­cast data col­lected reg­u­larly over time. So if you have annual prof­its, or monthly sales data, or weekly elec­tric­ity demand, or daily pas­sen­ger num­bers, or some­thing else col­lected reg­u­larly over time, then this course should be help­ful. But if your fore­cast­ing prob­lem does not fit into that par­a­digm, then it is prob­a­bly not the course you need.

Related Posts:

  • Pingback: Hyndsight - Online course on forecasting using R()

  • Raf­fael Vogler

    The two essen­tial ques­tions for me to decide whether this course is worth 1000 is:  1) Are there going to be exercises and will those be corrected by your staff? 2) Can you earn a certificate?  I am currently following a couple of courses about ML and statistics offered by Standford and Caltech - the lectures are AWESOME, the exercises are great and make sense und in some cases you can earn a certificate by getting enough points in exercises and final exam.  And it costs -0. I know this for-​​free-​​attitude is prob­lem­atic — I am will­ing to pay money on ser­vice — but $1000 is a lot of money and there has to be a good rea­son to pay that — espe­cially con­sid­er­ing the mar­ket situation.

    To make a long story short — I would love to fol­low this lec­ture in near future on edx​.org or cours​era​.org!

    Kind regards


    • Rob J Hyndman

      1. Yes there will be exer­cises. No they won’t be marked by me or my staff. We will dis­cuss the solu­tions in the online classes. I will prob­a­bly post solu­tions to class members.

      2. Monash Uni will not pro­vide a cer­tifi­cate. I sus­pect Rev­o­lu­tion Ana­lyt­ics would be happy to pro­vide a cer­tifi­cate of par­tic­i­pa­tion. I’ll ask them.

      I can­not use edX or Cours­era because they only take courses from uni­ver­sity affil­i­ated with them. My uni­ver­sity has just joined the Future Learn plat­form, so it is pos­si­ble that the course will be offered there one day, but there are no cur­rent plans to do so.

      In prin­ci­ple, any­thing free is being spon­sored by some­one. My text­book is spon­sored by OTexts​.com, and my R pack­ages are spon­sored by me and CRAN. Unless some­one is will­ing to spon­sor an online course on fore­cast­ing, it is going to cost something.

      Unfor­tu­nately, I don’t know of any free online courses on fore­cast­ing that I can recommend.

      • Raf­fael Vogler

        Thanks for sug­gest­ing the Future Learn plat­form and OTexts. I will have a close look at both — espe­cially your book on forecasting!

      • ilyakip­nis

        Well, on Cours­era, Eric Zivot has an intro/​review course on econo­met­rics, which includes some time series, and it’s all in R.

        • Rob J Hyndman

          I didn’t know that. I’m sure Eric will give a great course. But I doubt he will say much about forecasting.

  • Tao Hong

    $1000 is not expen­sive for a course taught by an inter­na­tion­ally rec­og­nized pro­fes­sor. How­ever, if I were run­ning Rev­o­lu­tion Ana­lyt­ics’ mar­ket­ing depart­ment, I would cover all the expenses you need and make this course for free to a broad audience.