Stella Kristín Hallgrímsdóttir

Benedikt Magnússon, May 29, 2017

Thesis presentation in Master of Applied Statistics (MAS)

Stella Kristín Hallgrímsdóttir
Title: Samband veðurs og komufjölda á bráðamóttökur Landspítala

Location: V-157, VRII
Time: Monday 29. May at 14:00.

Abstract:

The objective of this project is to study the seasonal and weekly fluctuations in number of arrivals to the emergency departments of the University Hospital of Iceland and also to assess the influence of weather on the number of arrivals. Four emergency departments were examined; the Emergency Department in Fossvogur, the Emergency Unit in the Children‘s Hospital Department, Hjartagátt which is the emergency department for people with suspected acute heart problems, and the Psychiatry Emergency Department. The weather variables that were mostly looked into are temperature, wind speed, precipitation and cloudiness. Seasonal fluctuations were modeled with sine and cosine waves and with the help of linear regression a new variable was made that describes the seasonal fluctuations and linear increase in the number of arrivals. A few ARIMA models were built to predict the number of arrivals in the Emergency Department in Fossvogur and in the Children‘s Emergency Department. The models were compared to find the best prediction model for each department. To assess whether weather affects the number of arrivals in the emergency departments, the weather variables were added one by one to the best prediction model for each department to see if the model‘s prediction root-mean squared error (RMSE) decreases when information about weather is added to the model. Principal components analysis was also used to combine the weather variables into fewer new variables. The new variables were then added to the ARIMA models to assess their effect on the goodness of the models. The results show that adding the weather information slightly decreases prediction RMSE in the Emergency Department in Fossvogur but increases it for the Children‘s Emergency Department. That both applies to when each weather variable was looked into separately and when the principal components were used. Therefore, it can be concluded that weather does not affect the number of arrivals to the Children‘s Emergency Department but it has a minor effect on the number of arrivals to the Emergency Department in Fossvogur. Furthermore, the results show that a good prediction model for the number of arrivals to the emergency departments can be developed only using calendar variables.

Advisors: Dr. Sigrún Helga Lund and Dr. Tryggvi Helgason
Examiner: Dr. Ólafur Pétur Pálsson

http://www.hi.is/vidburdir/meistarafyrirlestur_samband_vedurs_og_komufjolda_a_bradamottokur_landspitala