dc.contributor.advisor |
Rahamataullaha Imana, E. Eica. Ema. |
|
dc.contributor.author |
Alshammari, Abdulrahman Obaid |
|
dc.date.accessioned |
2015-05-14T12:28:04Z |
|
dc.date.available |
2015-05-14T12:28:04Z |
|
dc.date.issued |
2015-05-02 |
|
dc.identifier.uri |
http://cardinalscholar.bsu.edu/handle/123456789/199545 |
|
dc.description.abstract |
Road accident is considered as one of the major problems in the Kingdom of Saudi
Arabia. This motivates us to do research on this particular area. In our research the prime
objective is to find the most appropriate models for analyzing Saudi Arabia road accident data.
Since Saudi Arabia has several regions we model the data for the entire country and also for the
different regions. It is more likely that Box-Jenkin’s integrated autoregressive moving average
(ARIMA) models should fit the data. But the existence of missing values for each variable makes
the analysis part complicated since the estimation of parameters in an ARIMA model does not
converge when observations are missing. As a remedy to this problem we estimate missing
observations first. We employ the expectation maximization (EM) algorithm for estimating the
missing values. But since our data are time series data, any simple EM algorithm is not
appropriate for them. Hence we consider robust EM and bootstrap algorithms to estimate the
missing values. A study based on cross validation determines which the missing value estimation
techniques are the best for these data. Since we study time series data we employ a variety of ARIMA models for fitting and forecasting the number of accidents in Saudi Arabia and based on
a series of graphical and analytical tests finally determine which ARIMA model(s) are the best. |
en_US |
dc.description.sponsorship |
Department of Mathematical Sciences |
|
dc.description.tableofcontents |
Estimation of missing values, cross validation and ARIMA models in time series -- Estimation of missing values by robust EM and bootstrap algorithm for Saudi Arabia road accident data -- Selection of ARIMA models for Saudi Arabia accidents data. |
|
dc.subject.lcsh |
Traffic accidents -- Saudi Arabia -- Mathematical models. |
|
dc.subject.lcsh |
Time-series analysis -- Mathematical models. |
|
dc.subject.lcsh |
Missing observations (Statistics) |
|
dc.subject.lcsh |
Box-Jenkins forecasting. |
|
dc.title |
Modeling road accident data of Saudi Arabia |
en_US |
dc.description.degree |
Thesis (M.S.) |
en_US |
dc.identifier.cardcat-url |
http://liblink.bsu.edu/catkey/1784570 |
|