1. Go to statistikkbanken website and have a proper look at the different variables that are available. Think about those that you find more interesting and state a hypothesis you can test using some of the models you have learned in this course. If you do not find any variables of interest, or you want to combine the data with some variables from different sources, feel free to complement and/or get the full set of variables from other public sources. Please specify where the variables are coming from.
2. State the hypothesis. The hypothesis might have something to do with finance, economics, or accounting, but they may also be completely unrelated to these areas (for example, some sociological issues, sports, church attendance…) Do not try to base the hypothesis on previous literature, just explain it properly and explain also which model you are going to use to test it. Specify the model and explain why you choose a certain methodology.
3. Import the variables to STATA. Do as much data processing as you need in order to be in a position to apply your chosen methodology.
4. Make a table with the descriptive statistics you find most relevant.
5. Check if your variables fulfil the assumptions of the econometric model of your choice. For this you will need, for example, to check modelling issues thorough diagnostic residual plots and other relevant tests.
6. Run your model and report your results. Explain whether your hypothesis is accepted or rejected. It does not matter for your grade whether you find significance or not. The important thing in this exercise is that you show your knowledge about the econometric tools used to test your hypothesis and the program STATA.
7. Explain how good your model describes the data (goodness of fit).
8. Try to improve your model by including/excluding some variables, transforming variables, using some indicator variables (like time dummies, city dummies, industry dummies, etc.) Repeat steps 5 and 7 and report.
9. Finally, choose the model that you find more satisfactory.
In your answer, please report just the results and your comments in the main body of the paper. In an appendix, include the list of all the commands you have entered into STATA.
This material may consist of step-by-step explanations on how to solve a problem or examples of proper writing, including the use of citations, references, bibliographies, and formatting. This material is made available for the sole purpose of studying and learning - misuse is strictly forbidden.Summary
The aim of the paper was to test the relationship between Guest Nights, Tourist Expenditure and Air Passengers number. Data for the period between Q1 2009 to Q4 2018 were obtained from Statistics Norway. The Ordinary Least Square method was used for data analysis. Three model were tested to determine these effects. The first model showed that there is statistically significant and positive impact of Tourist Expenditure on Guest Nights, while there is no statistically significant impact of Air Passenger number on Guest Nights. Therefore, we transformed all variables into logarithmic form and that estimate the model. The results showed statistically significant impact of bot Tourist expenditure and Air Passengers number on Guest Nights, only the impact of Air Passengers is statistically significant at 10% significance level. At the end, we add to the second model dummy variable indicating high season. The results showed statistically significant impact of Tourist Expenditure and Season on Guest Nights. The most satisfactory model is the second model because both variables are statistically significant, although one at 10% significance level.
Data and Methodology
The following hypothesis will be tested: Tourist expenditure and number of air passengers have statistically significant and positive impact on guest nights.
Based on the set hypothesis, the following model was set:
(1) Guest Nightst = ∫ (Tourist Expendituret, Air Passengerst)
(2) Guest Nightst = β0 + β1*Tourist Expendituret + β2*Air Passengerst + εt
where: Guest Nights is number...