Micro-Simulation Modelling of Domestic Tourism in Sweden KIRUNA

Micro-Simulation Modelling of Domestic Tourism in Sweden KIRUNA Anders Lundgren Dep. Geography, Umeå University

SVERIGE System for Visualizing Economic and Regional Influences Governing the Environment SVERIGE is a Microsimulation model based on microdata on the Swedish population Socioeconomic data for ALL individuals

Current Modules Fertility Education Cohabitation Employment and marriage and earnings Mortality Immigration Divorce Leaving home Emigration Migration

What is microsimulation? • • Micro = individual level Microsimulation model - Simulation model that describes the behaviour of individual micro units and permits analysis of the impacts of the environment Laboratory ”world” for testing policies SVERIGE is dynamic and spatial - What happens if we increase immigration?

Experiment Base case – 60 000 immigrants 80 000 immigrants

”National” Tourism module • How many trips are done? • What do they do? • Where do people go? • Exchange of tourists between LAregions

Swedish Tourist Database • • Managed by Åre marknadsfakta 14 years 24 000 interviews every year Using 10 years

Number of cases • Randomly collected • Densely populated areas are well represented

Different categories of tourism Domestic travel with at least one night away from home done at leisure time

Limitations in data To few observations to perform regressions analyses IF you look at municipalities and each activity in TDB

LA-regions

Activity/purpose in TDB • Meeting friends and relatives • Visit second home • Piece and quiet/relaxation (Experience) • Pleasure and entertainment (Experience) • Community with others (Experience) • Skiing (Participate/be active in) • Sun&swimming (Participate/be active in) • Events (away 00 -01) • Outdoor life (Participate/be active in) • Sports (Participate/be active in) • Course & meeting as leisure assignment • Cultural activity (Participate/be active in) • Cultural environment (Experience) • Rush and speed (Experience) • Seclusion (Experience) • Prophylaxis, health care (Participate/be active in) • Natural environment (Experience) • Fishing (Participate/be active in) • Other activity (Participate/be active in) • Private matters/look for job (Experience) • See the country (Experience) • Buy things • Attraction (Visit, watch, listen to) • Stimulation (Experience) • Education/studies (Participate/be active in) • Urban environment (Experience) • Parks (Visit, watch, listen to) • Hunting (Participate/be active in) • Golf (Participate/be active in) • Adventure and excitement (Experience) • School trip (Experience)

Aggregated activities

A model for number of trips The two key factors that make tourism possible is access to money and leisure time Age group– divided into 5 groups Income – household income Gender Education – university degree or not If the individual has children at home or not If the individual is single or not Size of place of residence

A model for choice of activities Age, gender, income and lifecycle are key factors that affect activity choice Age group– divided into 5 groups Income – household income Gender Education – university degree or not If the individual has children at home or not If the individual is single or not Main residential region

Regression analysis Poisson regression for the number of trips – most people make 1 trip per month - 98% within 5 trips. Multinomial logit regression for the choice between 10 activities

Choice of destination • Interaction model used for destination choice • Compare calculated and observed number of trips from TDB • Interaction model estimated by using iteration and minimising misplaced flows

Results X 1000 • Distance is important for the activities VSH and sun/bath – people prefer to do that close to home. • Skiers and people who visits friends and relatives are less concerned about distance.

Tourism Flows Misplaced flows

Social Bond Activity

Visit Second Home

Sun and Bath

Skiing

Problems • Multinomial logit regression for destination choice failed • ”Empty” LA-regions – no observations • ”Holes” in the data – system missing

Data needs • More observations for individuals in sparsely populated areas • Purpose NOT mixed with activity • ”Clean” the array of variables with respect to desired information • What about immigrants? • Improve ”visit second home” and ”social bond” activities by using SCBdata

Future experiments • Change in demand by running ”SVERIGE” – 50 years forward – divorce, labour market, migration • Change supply – move central point of attraction

Summary • We model the flow of tourists between LA-regions • TDB explain choice of activities • An interaction model calculates the choice of place

END Questions?