Problem nejednakoga regionalnog razvoja jedan je od ključnih problema hrvatske ekonomske politike. Zakonom o regionalnom razvoju Republike Hrvatske uređeno je ocjenjivanje i razvrstavanje jedinica lokalne i regionalne samouprave prema razvijenosti, koja se temelji na indeksu razvijenosti. Ova univarijatna klasifikacija počiva na vrijednosti samo jedne varijable, koja je linearna kombinacija standardiziranih vrijednosti pokazatelja, što uzrokuje gubitak informacija. Cilj je ovoga rada prikazati alternativni pristup ocjenjivanju i kategorizaciji teritorijalnih jedinica prema socioekonomskoj razvijenosti na temelju multivarijatne analize. U radu je predočena klasifikacija jedinica lokalne i regionalne samouprave dobivena primjenom multivarijatnih metoda na temelju pet socioekonomskih pokazatelja. Poseban naglasak stavljen je na područja koja zaostaju u razvoju. Uočeni su i otklonjeni problemi multikolinearnosti i problemi postojanja izdvojenica koji su prisutni u klasifikaciji na temelju indeksa razvijenosti. Rezultati analiza upućuju na postojanje velikih razlika u regionalnoj razvijenosti. Gotovo sve metode klasterske analize rezultiraju posebnim klasterom u kojem su izdvojene jedinice koje imaju izražen problem depopulacije, ali ne i ekonomske nerazvijenosti, što omogućuje da se jedinice obuhvate posebnim mjerama poticanja demografskoga razvitka.

The development level assessment and categorization of Croatian local and regional units is based on the value of the development index which is the main instrument of Croatian regional policy. The development index is a composite indicator calculated as a weighted average of five socioeconomic indicators. The goal of this paper is to analyze the uncertainty and sensitivity of the development index that arise from the procedures and indicators used in its construction. This analysis is then used to propose useful guidelines for future impovements. The methodology of the Croatian regional development index has been critically reviewed, revealing problems of multicollinearity and the existence of outliers. An empirical and relatively more objective multivariate approach for weight selection has been proposed. The uncertainty and sensitivity analysis were conducted using Monte Carlo simulations and variance-based techniques. Instead of a unique point estimate for the development level of territorial units an alternative confidence interval approach was considered.

No abstract

As markets are becoming increasingly saturated, many businesses are shifting their focus to customer retention. In their need to understand and predict future customer behavior, businesses across sectors are adopting data-driven business intelligence to deal with churn prediction. A good example of this approach to retention management is the mobile game industry. This business sector usually relies on a considerable amount of behavioral telemetry data that allows them to understand how users interact with games. This high-resolution information enables game companies to develop and adopt accurate models for detecting customers with a high attrition propensity. This paper focuses on building a churn prediction model for the mobile gaming market by utilizing logistic regression analysis in the extended recency, frequency and monetary (RFM) framework. The model relies on a large set of raw telemetry data that was transformed into interpretable game-independent features. Robust statistical measures and dominance analysis were applied in order to assess feature importance. Established features are used to develop a logistic model for churn prediction and to classify potential churners in a population of users, regardless of their lifetime.

No abstract