What to cite?

This ques­tion comes from a com­ment on another post:

I’ve seen authors cit­ing as many ref­er­ences as pos­si­ble to try to please poten­tial ref­er­ees. Many of those ref­er­ences are low qual­ity papers though. Any gen­eral guid­ance about a typ­i­cal length for the ref­er­ence section?

It depends on the sub­ject and style of the paper. I’ve writ­ten a paper with over 900 cita­tions, but that was a review of time series fore­cast­ing over a 25 year period, and so it had to include a lot of references.

I’ve also writ­ten a paper with just four cita­tions. As it was a com­men­tary, it did not need a lot of con­tex­tual information.

Rather than pro­vide guid­ance on the length of the ref­er­ence sec­tion, I think it is bet­ter to fol­low some gen­eral prin­ci­ples of cita­tion in research. Con­tinue reading →

Dark themes for writing

I spend much of my day sit­ting in front of a screen, cod­ing or writ­ing. To limit the strain on my eyes, I use a dark theme as much as pos­si­ble. That is, I write with light col­ored text on a dark back­ground. I don’t know why this is not the default in more soft­ware as it makes a big dif­fer­ence after a few hours of writing.

Most of the time, I am writ­ing using either Sub­lime Text, RStu­dio or TeX­studio. Each of them can be set to use a dark theme with syn­tax col­or­ing to high­light struc­tural fea­tures in the text.
Con­tinue reading →

biblatex for statisticians

I am now using bibla­tex for all my bib­li­o­graphic work as it seems to have devel­oped enough to be sta­ble and reli­able. The big advan­tage of bibla­tex is that it is easy to for­mat the bib­li­og­ra­phy to con­form to spe­cific jour­nal or pub­lisher styles. It is also pos­si­ble to have struc­tured bib­li­ogra­phies (e.g., divided into sec­tions: books, papers, R pack­ages, etc.) Con­tinue reading →

Managing research ideas

I received this email today:

Dear Pro­fes­sor Hyn­d­man,
I was won­der­ing if you could maybe give me some advice on how to orga­nize your research process. I am able to search the lit­er­a­ture on a cer­tain topic and iden­tify where there is a ques­tion to work with. My main dif­fi­cult is to orga­nize my paper anno­ta­tions in order to help me to guide my research process, i.e, how to man­age the infor­ma­tion gath­ered in those papers to com­pose and struc­ture a doc­u­ment which can rep­re­sent the research devel­oped so far.
I have been look­ing at dif­fer­ent tools such scrivener, Qiqqa, papers2, etc but I am not sure if one of these tools would be the right way to go. To be hon­est I am not even sure a tool would do what I am look­ing for, not just orga­nize ref­er­ences and anno­tate pdfs but to get more con­trol of my research process.
I appre­ci­ate if I could get your thoughts on this subject.

Con­tinue reading →

Put your pre-​​prints online

I have argued pre­vi­ously that research papers should be posted online at the same time as they are sub­mit­ted to a jour­nal. Some­times peo­ple claim that jour­nals don’t allow it, which is non­sense. Almost every jour­nal allows it, and many also allow the pub­lished ver­sion of a paper to appear on your per­sonal website.

Today I dis­cov­ered a new tool (thanks to the IMU newslet­ter) which makes it easy to check a journal’s pol­icy on this. Check out SHERPA/​RoMEO.

It’s a very use­ful tool, but who­ever thought SHERPA/​RoMEO was a good name needs therapy.

My new forecasting textbook

After years of say­ing that I was going to write a book to replace Makri­dakis, Wheel­wright and Hyn­d­man (1998), I’m finally ready to make an announcement!

My new book is Fore­cast­ing: prin­ci­ples and prac­tice, co-​​authored with George Athana­sopou­los. It is avail­able online and free-​​of-​​charge. We have writ­ten about 23 of the book so far (all of which is already avail­able online), and we plan to fin­ish it by the end of 2012. We hope to make a print ver­sion of the book avail­able on Ama­zon in early 2013.

This text­book is intended to pro­vide a com­pre­hen­sive intro­duc­tion to fore­cast­ing meth­ods and present enough infor­ma­tion about each method for read­ers to use them sen­si­bly. We don’t attempt to give a thor­ough dis­cus­sion of the the­o­ret­i­cal details behind each method, although the ref­er­ences at the end of each chap­ter will fill in many of those details. We use R through­out the book and we intend stu­dents to learn how to fore­cast with R.

The book has it’s own R pack­age: fpp. This con­tains all the data sets used in the book, and also loads a few other pack­ages that are nec­es­sary to com­plete the examples.

The book is dif­fer­ent from other fore­cast­ing text­books in sev­eral ways.

  • It is free and online, mak­ing it acces­si­ble to a wide audience.
  • It is based around the fore­cast pack­age for R.
  • It is con­tin­u­ously updated. You don’t have to wait until the next edi­tion for errors to be removed or new meth­ods to be dis­cussed. We will update the book frequently.
  • There are dozens of real data exam­ples taken from our own con­sult­ing prac­tice. We have worked with hun­dreds of busi­nesses and orga­ni­za­tions help­ing them with fore­cast­ing issues, and this expe­ri­ence has con­tributed directly to many of the exam­ples given here, as well as guid­ing our gen­eral phi­los­o­phy of forecasting.
  • We empha­sise graph­i­cal meth­ods more than most fore­cast­ers. We use graphs to explore the data, analyse the valid­ity of the mod­els fit­ted and present the fore­cast­ing results.


Table design

Almost every research paper and the­sis in sta­tis­tics con­tains at least some tables, yet stu­dents are rarely taught how to make good tables. While the prin­ci­ples of good graph­ics are slowly becom­ing part of a sta­tis­ti­cal edu­ca­tion (although not an econo­met­rics edu­ca­tion!), the prin­ci­ples of good tables are often ignored. Per­haps peo­ple think they are obvi­ous, although the results I see in papers and the­ses sug­gest oth­er­wise. Con­tinue reading →