This thesis consists of three empirical studies of applied economics, in which different methodologies, techniques and statistical methods are used, as well as different sources of information, among which are traditional and non-traditional sources of data. The first chapter aims to understand the relationship between differentiation and resilience for the formulation and implementation of public and private initiative measures that allow early identification of risks to minimize the fragility of certain housing markets. In the hotel sector there is a high degree of horizontal differentiation, which can occur in the location or the offer of services that is adapted for the different types of clients. This evaluates the heterogeneous effects that occur in the face of exogenous falls in demand, according to the level of horizontal differentiation. For this, different data sources are used that are integrated, such as: i) Administrative Data, ii) surveys, and iii) non-traditional sources of information, obtained through searches on web portals that facilitate the sale of hotel rooms. and the Google API. Based on the estimation of differences-in-differences specifications, the results suggest a greater resilience in those establishments that make a greater effort to differentiate themselves geographically or in their service offer. The second chapter aims to study the impact of the temporary change in the data collection of the Great Integrated Household Survey (GEIH), from which information related to income and poverty calculations is obtained, as a consequence of the sanitary measures implemented by the National Government of Colombia between March and July 2020. The strict confinement implemented by the National Government as a measure to contain the spread of the pandemic caused by COVID-19 generated challenges in data collection operations through household surveys . As a result, the surveys with face-to-face collection methods migrated to a remote modality, through telephone surveys, which could have modified the possible reporting biases of variables such as income. This article studies the effect of a change in the information collection model in the Great Integrated Household Survey (GEIH) of Colombia on the report of labor income. To do this, the geographic variation in the implementation of collection methods and the integration of the survey with a social security administrative record are exploited to quantify the variation in the report. The results are relevant to study the impact of collection methods on survey reporting biases and the use of data integration methods from different sources. This chapter was co-authored with Juan Daniel Oviedo, Andrés García Suaza and José Lobo. Finally, the third chapter aims to analyze the way in which online shopping platforms influence the decision making of travelers when choosing hotels, as well as the effects on their profitability and that of their competition. The growing use of platforms for online purchases of experiential goods whose purchase is not recurring and in which the supplier has information privileges, such as accommodation, generates positive externalities, since past consumer experiences provide useful information for decision-making. decisions by travelers and competing hotels. In this way, this article will demonstrate, through a Durbin autoregressive spatial model, how the disclosure of information about past experiences in competing hotels with similar attributes generates an effect on own profitability. This research uses an innovative database that links data from accommodation surveys and tourism administrative records with data collected by a web-crawler from the facilitating web platforms and the Google Cloud API
