The purpose of this thesis is to investigate the possibilities of predicting vacancies in the real estate market by using machine learning models in terms of classification. These models were mainly based on data from contracts between a Swedish real estate company and their tenants. Attributes such as annual renting cost and rental area for each contract were supplemented with additional data regarding financial and geographical information about the tenants. The data was stored in three different formats with the first having binary classes which aim is to predict if the tenant is moving out within a year or more. The format of the second and third version were both multi classification problems that aims to classify if the tenants might terminate their contract within a specific interval with the length of three and six months. Based on the results from Microsoft Azure Machine Learning Studio, it is discovered that the multi classification problems perform rather poorly due to the classes being unbalanced. Regarding the  performance of the binary model, a more satisfying result was obtained but not to the extend to say that the model can be used to determine a vacancy with high accuracy. It should rather be used as a risk analysis tool to detect if a tenant is showing tendencies that could result in a future vacancy. A major pitfall of this thesis was the lack of data and the financial information not being specific enough. The performance of the models will likely increase with a larger dataset and more accurate financial information.

The Stockholm office market segment have for a long time been considered a safe haven when it comes to withstand negative turmoil in the form of rental compression due to economic fluctuation, especially in the CBD demographic. Recently however, a large number of banks and institutions, amounting to some 200000 square meters, have decided to relocate to more peripheral locations with the aim of cost reductions on rent. This mass exodus is studied with focus on rental dynamics as the result of increased vacancies. Other variables are stock changes and employment. The method is econometric combined with an interview series. The data is a panel dataset containing 900 observations. The different models that are being used is the vacancy gap model, an Error Correction Model and a dynamic lag model in the form of a first difference model. Due to non-stationary variables, some models were rejected. A prognosis model has been created for the economic calculations. The results are displayed in a number of scenarios ranging from unchanged rents to severe rental drops. As a comprehensive result this study concludes that a rental drop in the range of 10-20 percent is to be expected. A number of positive side effects are expected to unfold as a result of the relocations. / Stockholmsmarknadens kontorssegment har länge setts på som säker vad gäller att kunna motstå negativ turbulens i form av hyres press på grund av ekonomiska svängningar, och då speciellt marknaden för City eller CBD. Nyligen har dock en stor mängd banker och institutioner om cirka 200000 kvadratmeter bestämt sig för att lokalisera sig i mer perifera lokaler med kostnads reduceringar i form av lägre hyror som mål. Den här massutflyttningen har studerats med fokus på hyresdynamik som ett resultat av ökade vakanser. Andra variabler är ändringar i stocken och sysselsättning. Metoden är ekonometrisk och kombineras med en intervjuserie. Data består av ett paneldatasett innehållande 900 observationer. De olika modellerna som används är en vakansgapsmodell, en Error Correction modell (ECM) och en första differens modell. På grund av icke stationära variabler har vissa modeller förkastats. En prognostiseringsmodell har skapats för de ekonomiska beräkningarna. Resultaten redovisas i olika scenarion som består i allt från att hyrorna inte ändras alls till svåra hyresfall. Ett samlat resultat av studien föreslår ett hyresfall på 10-20 procent. En mängd positiva bieffekter förväntas uppstå som ett resultat av om lokalisationerna.