@VMihova
Ruse,
Membre depuis le 9 août 2018
0 Recommendations

# VMihova

En ligne Hors ligne
PhD in Mathematical Modeling. Dissertation topic: "A Study on Econometric Models in Finance". Previous experience as: - Solutions Analyst at Experian Bulgaria (data preparation with statistical software, analysis and modeling of credit risk) - Part-time Lecturer at Sofia University, Faculty of Mathematics and Informatics (Simple Differential Equations, Statistics and Empirical Methods) - Assistant Professor at University of Ruse, Department of Applied Mathematics and Statistics (Applied Statistics, Optimal Control and Operations Research, Econometrics).
\$10 USD/hr
7 commentaires
3.7
• 100%Travaux complétés
• 78%Selon le budget
• 78%Dans les temps
• 100%Taux de Réembauche

## Expérience

### Assistant Professor

Oct 2013 - Oct 2017 (4 years)

Teaching students in Applied Statistics, Optimal Control and Operations Research, Econometrics.

### Analyst and Solutions Analyst

Nov 2009 - Jan 2013 (3 years)

Data Preparation and Credit Risk Modeling and Analysis, supporting junior analysts and training new starters.

### Part-time lecturer

Oct 2009 - Jun 2010 (8 months)

Teaching students in Simple Differential Equations and Statistics and Empirical Methods.

### Reporter

Jun 2007 - Sep 2009 (2 years)

Writing articles on various economic topics; interviews; polls; reflecting press conferences.

## Éducation

### PhD

2016 - 2018 (2 years)

### Master

2009 - 2013 (4 years)

### Bachelor

2005 - 2009 (4 years)

## Publications

### Comparative analysis on the probability of being a good payer

This work studies the causal link between the conduct of an applicant upon payment of the loan and the data that he completed at the time of application. A database of 100 borrowers from a commercial bank is used. Customers are divided into Good and Bad payers, based on the credit history. Linear and logistic regression are applied in parallel to the data in order to estimate the probability of being good for new borrowers. A comparative analysis of the results is made.

Online advertising services such as Google AdWords provide their customers with statistics on the performance of their ads. To optimize the advertising strategy, various mathematical tools could be used, one of which – the decision tree. In the presented work, by solving a problem from practice, the potentials of using this tool for classification in the field of online advertising are outlined.

### An Approach Of Estimating The Probability Of Being Good For New Borrowers

This work has studied the causal link between the conduct of an applicant upon payment of the loan and the data that he completed at the time of application. A linear regression is used to estimate the probability of being good for new borrowers, and a scorecard is obtained from the linear model to assess new customers in the time of application.

### A classification of borrowers from commercial banks (in Bulgarian)

A common tool to improve competitiveness in the banking sector is offering a special range of products targeting loyal customers or the so-called Loyalty program. The aim of this work is to show in practice how could customers be segmented with cluster analysis. In this context an example of the grouping of loyal customers in separate groups ("platinum", "gold", "silver", etc.), using cluster analysis is considered.

## Vérifications

• Freelance Préféré
• Paiement vérifié
• Téléphone Vérifié
• Identité Vérifiée
• E-mail vérifié

• 7
• 7
• 7
• 7
• 7