Research tips

Rob J Hyndman: Research tips

Posts tagged references

Econometrics and R

31Aug 2010

Posted by Rob J Hyndman in Uncategorized | 2 Comments

Econo­me­tri­cians seem to be rather slow to adopt new meth­ods and new tech­nol­ogy (com­pared to other areas of sta­tis­tics), but slowly the use of R is spread­ing. I’m now receiv­ing requests for ref­er­ences show­ing how to use R in econo­met­rics, and so I thought it might be help­ful to post a few sug­ges­tions here.

A use­ful on-line and free resource is “Econo­met­rics in R” by Grant Farnsworth. It cov­ers some com­mon econo­met­ric meth­ods includ­ing het­eroskedas­tic­ity in regres­sion, pro­bit and logit mod­els, tobit regres­sion, and quan­tile regres­sion. In the time series area, it cov­ers ARIMA, ARFIMA, ARCH and GARCH mod­els, as well as a few of the stan­dard tests for unit roots and auto­cor­re­la­tion. It’s brief but it does pro­vide code that will help peo­ple famil­iar with econo­met­rics to get started using R.
If you are pre­pared to pay, an excel­lent book is Kleiber and Zeilis’s Applied Econo­met­rics with R. It cov­ers sim­i­lar ground to Farnsworth but in more detail. This is the book I usu­ally rec­om­mend to any­one with an econo­met­rics back­ground who is want­ing to get started with R. It would also be very suit­able for some­one study­ing econo­met­rics at about upper under­grad­u­ate level. Achim Zeileis is a well-known expert in R pro­gram­ming, so you can be sure the code in this book is effi­cient and well-written.
Another use­ful book is Pfaff’s Analy­sis of Inte­grated and Coin­te­grated Time Series with R which cov­ers unit root tests, coin­te­gra­tion, VECM mod­els, etc.
Vinod’s Hands-On Inter­me­di­ate Econo­met­rics Using R con­tains a lot of exam­ples and code-snippets which can be very help­ful. Unfor­tu­nately, the exam­ples do not always show the best prac­tice in R coding.
More detailed case stud­ies using R are pro­vided in Advances in Social Sci­ence Research Using R, edited by H.D. Vinod. Many of the case stud­ies are from econo­met­rics includ­ing an excel­lent chap­ter by Bruce McCul­lough on econo­met­ric computing.

There are of course dozens of books on R with a more sta­tis­ti­cal per­spec­tive, includ­ing sev­eral on time series. But I will leave them for another post.

  • Share/Bookmark

Tags: ,

Academic citations in the popular press

12Jul 2010

Posted by Rob J Hyndman in Uncategorized | No Comments

It is very unusual for a news­pa­per arti­cle to cite an aca­d­e­mic paper, unless it is in Nature, Sci­ence or the Lancet. Mostly, what we write is too tech­ni­cal and assumes too much back­ground knowl­edge for it to be acces­si­ble to any­one but spe­cial­ists. So I was pleas­antly sur­prised to find a ref­er­ence to the Inter­na­tional Jour­nal of Fore­cast­ing in a recent Wall Street Jour­nal arti­cle. It is a cita­tion of a 1996 arti­cle, so in terms of sci­en­tific research it is a bit like quot­ing the Magna Carta, but a cita­tion nevertheless.

I once tried to get news­pa­per cov­er­age of a spe­cial issue of the IJF on fore­cast­ing the US Pres­i­den­tial elec­tion. It was pub­lished about four months before the 2008 elec­tions. If any­thing was going to attract the atten­tion of the pop­u­lar press, surely this was the topic! Alas, all we man­aged was a short piece on a  research news web­site although there were copi­ous arti­cles on pre­dict­ing the elec­tion result based on less valid methods.

Even fore­cast­ing the recent world cup didn’t get any seri­ous atten­tion, despite some excel­lent (albeit unpub­lished) work over at kaggle.com. Paul the Octo­pus had tens of thou­sands of news arti­cles, but the care­ful sta­tis­ti­cal mod­el­ling at kag­gle had none at all that I could find.

All of which goes to show that news­pa­pers are not good sources of infor­ma­tion about fore­cast­ing (or any­thing else?).

  • Share/Bookmark

Tags: ,

Google scholar alerts

26May 2010

Posted by Rob J Hyndman in Research tips | No Comments

A cou­ple of weeks ago, Google scholar added a facil­ity to pro­vide email alerts on new arti­cles asso­ci­ated with spe­cific search queries. First do the search, then click the enve­lope at top left of screen. For exam­ple, here is a search on “expo­nen­tial smooth­ing” since 2000.

Note the enve­lope at the top marked New! Click it to get the fol­low­ing screen.

Those results show some of the flaws in Google Scholar — the dates are not always cor­rect (the first paper listed above appeared in 2004) and there are unre­solved duplicates.

Despite the prob­lems, if you’re want­ing to keep an eye out for new papers on par­tic­u­lar top­ics, this looks like it could be use­ful. Unfor­tu­nately, there is no RSS feed available.

  • Share/Bookmark

Tags: , ,

Using DOIs

27Dec 2009

Posted by Rob J Hyndman in Research tips | 1 Comment

Almost all papers these days have a DOI and it is worth know­ing how to use them.

At the top or bot­tom of the first page of a paper, you will see some­thing like this:

doi:10.1016/j.csda.2006.07.028

This is a unique and per­ma­nent iden­ti­fier for the paper known as a “Dig­i­tal Object Iden­ti­fier”. The part before the for­ward slash (10.1016 in the exam­ple above) iden­ti­fies the nam­ing author­ity (in this case Else­vier) and the part after the for­ward slash (j.csda.2006.07.028) iden­ti­fies the par­tic­u­lar paper. In this case, the paper iden­ti­fier shows it is in the jour­nal Com­pu­ta­tional Sta­tis­tics and Data Analy­sis and that it first appeared online in 2006. How­ever, there is no sys­tem­atic pat­tern to these iden­ti­fiers, and other jour­nals use other ways of gen­er­at­ing identifiers.

One use for these num­bers is that it pro­vides a quick way of find­ing the paper online. The URL http://dx.doi.org/xxx where xxx is the DOI will lead to the paper. For exam­ple, http://dx.doi.org/10.1016/j.csda.2006.07.028 gives the above paper.

A URL gen­er­ated in this way is usu­ally much shorter than other equiv­a­lent URLs, and is guar­an­teed not to change, even when the pub­lisher reor­ga­nizes their web­site. For some jour­nals, the URL of an arti­cle may change when the arti­cle moves from being online but not yet allo­cated to an issue, to being part of a print issue. But the DOI will remain the same regard­less. That is why I usu­ally pro­vide links of this form to my online pub­li­ca­tions from my website.

  • Share/Bookmark

Tags: ,

“Elements of Statistical Learning” now online

15Oct 2009

Posted by Rob J Hyndman in Research tips | 1 Comment

In the past cou­ple of days, the authors of sev­eral blogs have noted that the won­der­ful book The Ele­ments of Sta­tis­ti­cal Learn­ing: Data Min­ing, Infer­ence, and Pre­dic­tion by Hastie, Tib­shi­rani and Fried­man (2nd ed., 2009) is now avail­able for free down­load in pdf for­mat.

Of course, it is also nice to have a hard copy. Click the image to pur­chase from Amazon.

  • Share/Bookmark

Tags:

Sight what you cite

13Aug 2009

Posted by Rob J Hyndman in Research tips | 1 Comment

There seems to be a wide­spread prac­tice of researchers cit­ing papers they have never even seen, let alone read. For example

  • Some papers claim to do some­thing new when it has already been done in one of the papers cited.
  • Some arti­cles are cited that appar­ently have lit­tle to do with the rea­son given for the cita­tion, or which argue the oppo­site point of view to what is claimed.
  • A paper uses method X and cites an arti­cle which points out a flaw in method X, with no com­ment on why the flaw can be ignored.
  • Ref­er­ence details are so full of errors that the paper can never have been sighted.
  • Some mis-spellings of an author’s names are per­pet­u­ated through long series of copied cita­tions with no-one ever check­ing the details. I fre­quently cor­rect the list of ref­er­ences in papers to appear in the Inter­na­tional Jour­nal of Fore­cast­ing, but only where I notice an error, or where it looks sus­pi­cous. I haven’t time to check them all; besides, that’s the author’s job.

William Web­ber points out that there are cited papers that have never even been writ­ten! One famous paper has 215 cita­tions on Google scholar despite the fact that it doesn’t exist. And at least one of those cita­tions is by the author of the non-existent paper!

How can this hap­pen? It is sim­ply sloppy, lazy, and unac­cept­able. Here are a few com­ments that I regard as the bare min­i­mum for respon­si­ble scholarship.

  1. Every arti­cle cited should be sighted, and prefer­ably read.
  2. At the very least, check that the arti­cle cited really does say what you think it says.
  3. Type the ref­er­ence infor­ma­tion your­self from the orig­i­nal source, not copied from some­one else’s citation.
  4. Don’t just cite what other peo­ple say about cita­tions. For exam­ple, “Paper 1 says x about Paper 2″. Unless you have read both Paper 1 and Paper 2, and you have some­thing to say about the state­ment x in Paper 1, there is no point sim­ply repeat­ing it. It makes you look lazy and inca­pable of form­ing your own opin­ion. If Paper 1 is incor­rect, then you look stu­pid as well.
  5. Avoid lists of gra­tu­itous ref­er­ences. For exam­ple, “There has been a lot of recent work (e.g., a, b, c, d, e, f, g) on this topic.”  Instead, cite the rel­e­vant back­ground lit­er­a­ture, and make com­ments about how each cited paper fits in with what your paper is about. Just list­ing papers for the sake of extend­ing your own bib­li­og­ra­phy is rather pointless.
  6. Before you sub­mit your paper, go through the bib­li­og­ra­phy and check the details one more time.
  • Share/Bookmark

Tags: ,

Managing a bibliographic database

2Aug 2009

Posted by Rob J Hyndman in Research tips | 5 Comments

All researchers need to main­tain a data­base of papers they have read, cited, or sim­ply noted for later ref­er­ence. For those of us using LaTeX, the data­base is in the Bib­TeX for­mat and is stored as a sim­ple text file (a bib file) that can be edited using a text edi­tor such as WinEdt.

But it is often eas­ier to edit the file using spe­cial­ist soft­ware. My cur­rent favourite tool is JabRef. It is espe­cially use­ful when search­ing for ref­er­ences in a large data­base. How­ever, there are sev­eral things I would like to be able to do with my bib file, but JabRef doesn’t pro­vide the facilities.

First, when I save a pdf of an arti­cle to my local disk, I would like the bib details to be auto­mat­i­cally added to my data­base. But I cur­rently have to go and type them in myself. Usu­ally I don’t bother until I need to cite the paper, but then I have trou­ble try­ing to find a paper that I’m sure I’ve saved, but can’t find. Even Google Desk­top seems to have trou­ble find­ing papers that I’ve saved.

Sec­ond, it would be nice to share my bib file with my research stu­dents and post-docs. Cur­rently I have to send them a copy periodically.

Third, when I find a paper on Sci­enceDi­rect, or RePEc, or some other online repos­i­tory, it would be great if I could just click a but­ton and have the paper added to my data­base, along with a link to the URL where the paper can be obtained.

In the last year, there has been a lot of work going on in try­ing to develop new tools to solve some of these prob­lems. There are now tools avail­able to do almost every­thing I want, but no one tool does them all. Unfor­tu­nately, many of the avail­able tools are designed to work with MS-Word (ugh) and don’t inte­grate neatly with LaTeX/BibTeX.

One of the most promis­ing of the new tools appears to be Mende­ley. I’ve tried import­ing my bib file into Mende­ley, but it seems to fall over with any­thing more than a few hun­dred items. My bib file con­tains about 3000 items. I’m told they are work­ing on per­for­mance issues for the next release, so I’ll look for­ward to try­ing it out again in a few months.

In the mean­time, I’ll stick with my tried and trusty JabRef, despite its limitations.

  • Share/Bookmark

Tags: , ,

Searching the research literature

14Jul 2009

Posted by Rob J Hyndman in Research tips | No Comments

Most stu­dents seem to go to Google first. This is not a good strat­egy. Google Scholar is much bet­ter as it fil­ters out all the junk. Sco­pus is another engine that aims to do a sim­i­lar thing. It is bet­ter orga­nized but not so com­plete. ISI WOK is also not as com­plete as Google Scholar but is par­tic­u­larly good at track­ing citations.

If you track down some­thing on Google scholar, but are at home and so do not have access to the paper, you can use Monash ezproxy. Just drag the pre­vi­ous link to your book­marks. When you are on the page of a paper you want, click the link and it will ask for your usual Monash username/password, after which you should be able to access the paper.

Some­times, a more restric­tive search is needed. Try one of these:

Finally, if you’ve still had no luck, then try Google.

If you know the paper you are after, it is often bet­ter to go through the Monash library. You can search for an elec­tronic jour­nal via the Monash library.

Notes:

  • A few of these links require a licence to access them. All of them can be accessed from Monash computers.
  • You don’t need to search JSTOR sep­a­rately as it is cov­ered by Google scholar.
  • RePEc is cov­ered by both Sco­pus and Google Scholar
  • Share/Bookmark

Tags: ,

Accessing journal articles online

12Mar 2009

Posted by Rob J Hyndman in Research tips | No Comments

When search­ing for research arti­cles online, I often find that the arti­cle is unavail­able unless I go through the Monash library web­site, espe­cially when work­ing from home. Here are two solu­tions to the problem

  1. Within Google scholar, go to “Scholar pref­er­ences” and under library links search for “Monash”. Tick the entry “Monash Uni­ver­sity — Check for full text”. Then save your pref­er­ences (but­ton at bot­tom of page). Next time you do a Google scholar search, a link labelled “Check for full text” will appear beside each entry. Click it and you will be able to eas­ily go straight to the arti­cle after enter­ing your Monash user­name and password.
  2. Drag the fol­low­ing book­mark to your book­marks tool­bar: Monash ezproxy.
    Then when­ever you are on a journal’s web­site and want to down­load an arti­cle, click the book­mark and it will ask for your Monash user­name and password.

Of course, this only works for Monash stu­dents and staff. But other read­ers should be able to adapt the above pro­ce­dures for other universities.

  • Share/Bookmark

Tags: , ,

RSS feeds for statistics journals

24Aug 2008

Posted by Rob J Hyndman in Research tips | No Comments

I am slowly adding more sta­tis­tics jour­nals to my RSS feed on new sta­tis­tics research papers. I am now cov­er­ing 62 jour­nals as well as work­ing papers lodged on the sta­tis­tics sec­tion of arXiv.  The most recent addi­tions are the sta­tis­tics jour­nals stored on Project Euclid, which includes the excel­lent IMS journals.

If your research inter­ests are close to mine, sub­scrib­ing to this feed will save you hav­ing to set up all the feeds yourself.

  • Share/Bookmark

Tags: , ,