Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Robert A Yaffee <bob.yaffee@nyu.edu> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: Binary time series |

Date |
Tue, 21 Sep 2010 23:22:26 -0400 |

John, Irregularly spaced time series has been handled by attempts to model intermittend demand. Croston's method and variations on it have been used to handle such data. Unfortunately, Stata has no command for Croston's method. In high frequency financial data, they attempt to model integrated volatility or realized volatility by taking the absolute or squared value of transactions at various time intervals. Sometimes the interval is a five or ten minute interval. Sometimes the interval is the complete workday, during which the value of the transactions might be summed. The problem with the square is that jumps in volatility may occur that can complicate the assessment of volatility if they are not taken into account. One thing to consider is how to decide upon the proper interval in which a variance can be computed. Croston's method uses a simple exponential smoother to handle the interval between observations. A different exponential smoother used to account for the magnitude of the observation. Finally, to achieve the mean demand rate, Croston divides the magnitude of demand by the interval time, each of which are the dependent variables in the simple exponential smoother. There have been modifications of this method by Johnston and Boylan and Boylan and Syntetos somewhat later. Siem Jan Koopman with others has set forth a generalized scoring algorithm handling such time series models. But I haven't seen the software yet. These may give you some ideas. Cheers, Robert Robert A. Yaffee, Ph.D. Research Professor Silver School of Social Work New York University Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf CV: http://homepages.nyu.edu/~ray1/vita.pdf ----- Original Message ----- From: John Morton <john.morton@optusnet.com.au> Date: Tuesday, September 21, 2010 8:14 pm Subject: st: Binary time series To: statalist@hsphsun2.harvard.edu > Hi again, > > Same message as 90 minutes ago, this time with a subject heading. My > apologies for overlooking this in the previous post. > > > I am seeking advice on analysis of a time series dataset in Stata. The > same > site was visited irregularly 30 times over 3 years (median interval between > visits 35 days, range 18 to 68 days). At each visit, usually 5 > tadpoles (but > sometimes 6 or 9) were sampled (numbers were limited because this is an > endangered species). Different tadpoles were sampled at each visit. Each > tadpole was tested and categorised as test positive or test negative. > Apparent prevalences were 1.00 at about half of the visits and 0.00 at > about > 25% of visits. > > The researcher?s question is whether prevalence varies by month (ie Jan, > Feb, Mar etc) or by season. > > The features of this data that seem important are that the errors > would be > expected to be serially correlation over time, the dependent variable > is > binary, prevalences of 0 and 1 were common, the very small number of > tadpoles sampled at each visit, and these are not panel data (ie different > tadpoles were sampled at each visit). > > I have done some exploratory modelling treating prevalence as a continuous > dependent variable (using -regress-) after declaring the data to be > time-series data (with sequential visit number rather than day number > as the > time variable, using -tsset-). With a null model, tests for serial > correlation (Durbin-Watson test (-estat dwatson-), Durbin?s > alternative (h) > test (-estat durbinalt-),Breush-Godfrey test ( -estat bgodfrey,lag(6)-), > Portmaneau (Q) test (-wntestq-) and the autocorrelogram (-ac-)(all > from Baum > 2006) indicate serial correlation. In contrast, after fitting month as > a > fixed effect, these tests do not support rejecting the null hypothesis > that > no serial correlation exists. However treating prevalence (a > proportion) as > a continuous dependent variable (using -regress-) is inappropriate. > > Any suggestions on approaches to answer the research question would be > much > appreciated. > > Many thanks for any help. > > John > > *************************************************************** > Dr John Morton BVSc (Hons) PhD MACVSc (Veterinary Epidemiology) > Veterinary Epidemiological Consultant > Jemora Pty Ltd > PO Box 2277 > Geelong 3220 > Victoria Australia > Ph: +61 (0)3 52 982 082 > Mob: 0407 092 558 > Email: john.morton@optusnet.com.au > *************************************************************** > > > > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/statalist/faq > * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**Re: st: Binary time series***From:*Robert A Yaffee <bob.yaffee@nyu.edu>

- Prev by Date:
**st: Number of observations reported by xtreg with fixed effects** - Next by Date:
**Re: st: Binary time series** - Previous by thread:
**st: irregularly spaced time series** - Next by thread:
**Re: st: Binary time series** - Index(es):