Data modelling, Analysis, Statistical Planning for loan market

$10-120 USD

Closed

Posted

almost 9 years ago

$10-120 USD

Paid on delivery

Gearcash app is a cash lending platform. We collect 5 basic fields for borrower profile: age, rented or bought house, annual income, number of loans, total debt. Utilising these 5 fields, you will provide me with.... 1) a formula to get a "Gearcash Basic Credit Score", where point 1 is the poorest credibility, and point 100 is the healthiest credibility. (Do your research on how typical credit scores are formulated and provide your own twist.). Provide me the technical explanantion to how you derive this score and explain in detail the calculation and why this is a more accurate way of determining the "Gearcash Basic Credit Score". apart from the above 5 profile fields, we will additional data from his usage of the app, we call this data "loan behavior". The data fields we will have will be: Loan amount requested, number of times requested, number of loans taken up from Gearcash, loan amount accepted, loan interest rate accepted. 2) i need another formula to get a "Gearcash Advanced Credit Score". This advanced credit score is basically the "Gearcash Basic Credit Score" PLUS his "loan behavior" data. So you would need to break this down into 2 parts a) Formulating a "loan behavior" score, where 1 is the poorest loan behavior, and 100 is the healthiest loan behavior. Provide me the technical explanantion to how you derive this score and explain in detail the calculation and why this is a more accurate way of determining the "loan behavior". b) Formulating the "Gearcash Advanced Credit Score", which is basically the "Gearcash Basic Credit Score" MERGED with his "loan behavior" data. Provide me the technical explanantion to how you derive this score and explain in detail the calculation and why this is a more accurate way of determining the "loan behavior". So in summary of above, i need 3 forumlas developed as above.

Project ID: 7771379

About the project

17 proposals

Remote project

Active 9 yrs ago

Looking to make some money?

Email address

Benefits of bidding on Freelancer

Set your budget and timeframe

Get paid for your work

Outline your proposal

It's free to sign up and bid on jobs

17 freelancers are bidding on average $102 USD for this job

@ACMAuk

Hi, I am Chartered Management Accountant from CIMA-UK as well as MSc Statistics with excellent hands on experience in MS Excel. We would be requiring more information to develop the model required, and we may be using Excel macros for that. Awaiting your response, Cheers, AHK

$100 USD in 5 days

5.0

(171 reviews)

6.8

@sheak

Quality writer at your service,please consider my bid for high quality work on time. Hi, Are you little worried and finding freelancer for your task, i am here to deliver beyond your expectations. Award me project without wasting your precious time and allow me to start right away, i am fully capable to complete your task within deadline. I assure you 100% quality work on time. I am a vigilant, hard-working, reliable and smart working writer who delivers project according to clients requirement. i am confident you will get best from me. hope you will consider me,thanks\

$100 USD in 3 days

4.8

(26 reviews)

6.3

@karanrajput1112

As a Post Graduate in Statistics and graduate in mathematics, i have a lot of experience in handling statistical data, especially in Time Series Modelling & Forecasting, Regression analysis,correlation , ANOVA, statistical hypothesis testing, Descriptive measure etc. I have done various projects based on statistical concepts with the help of different statistical software packages like R, SPSS, Minitab etc. You will highly appreciate my work if you give me a chance to do your work professionally

$142 USD in 1 day

4.9

(49 reviews)

5.5

@d2imatrix

Hi, We are credit risk analysts and have been working on credit scoring for last 10 years. Here is what we can provide you. 1. A logistics regression based model to get the parameter estimates first for the "Gearcash Basic Credit Score" model. This will be done using fine-classing of variables, then coarse classing to ensure the linearity, then finally the logistic regression model. Later we will scale the probability into a score which will range between 1 to 100. Higher the score greater credit worthiness. Finally you will get the scorecard equation to put into any software engine and get predictive scores. 2. There are two ways we can build the second model. a) Include earlier 5 variables and new behavioral variables. Then start doing the modelling to get "Gearcash Advanced Credit Score". Advantage of this is the new model will not have any direct dependency on the first "Gearcash Basic Credit Score" model. b) Use "Gearcash Basic Credit Score" as an independent variable and also include new behavioral variables to create "Gearcash Advanced Credit Score". We will do this in SAS. We can really do this job, however the amount you have mentioned is pretty less for these kind of models. Kindly let us know if you want to discuss these. Regards, Pal

$120 USD in 5 days

4.6

(19 reviews)

4.9

@ashurao

~~~Graduate in Applied Statistics~~~ Expert: @Time Series Analysis, Econometrics, Design of Experiment, @Multivariate analysis,Linear Regression analysis, @Bio-Statistics, Actuarial Statistics. Programming (Data analysis tools): # R, SPSS, SAS, STATA, Excel. # Data interpretation and Reporting (using Latex). #100% Dedication for customer satisfaction.

$100 USD in 1 day

4.9

(5 reviews)

3.2

@sjainfirst

I have a good experience of working with various lenders for debt syndication and believe I can help with this project.

$90 USD in 5 days

0.0

(0 reviews)

0.0

@farhan00100

I am a qualified Professional, who initiates sophisticated assignments, meets tight deadline and brings out quality performance. Possess practical knowledge and experience in corporate finance and accounting practice. Implement substantial financial planning and analytical proficiency to assess the effectiveness of the current resources. Able to produce management reports to distinguish the overall profitability. A chartered accountant, experienced and motivated contender worked in various industries. Furthermore, having experience to conduct full scope audits as well as dealing with the auditors and manage the proceedings quite efficiently which help me to identify enormous findings during the task. I'd appreciate to discuss this opportunity with you and looking forward to hear from you. Regards, Farhan Farooq Patel.

$77 USD in 4 days

0.0

(0 reviews)

0.0

@Amritpratihasta

Dear Sir, I am a banker, with a huge interest in analytics. I find your project really interesting. I have excellent excel skills. I have done several projects on SAS. Looking forward to work for your project.. All I want in return is a good rating and blessings from you. Regards, Amrit Kumarr Pratihasta

$10 USD in 10 days

0.0

(0 reviews)

0.0

@edwardsjeremy

A proposal has not yet been provided

$111 USD in 1 day

0.0

(0 reviews)

0.0

@helena25h

This is an interesting task that requires creativity. Technicality it will include analysis of at least two datasets and lots of documentation writing. This project requires a unique approach and hence requires time to complete it in a right way. It is important to know the size of your datasets (how many rows), and what credibility (task 1) and loan behavior (task 2) means to you. Is a good person a person who borrows more money? or is it the one who lets you earn more? Could you mark the rows of data you have with a credibility and behavior scores? That way formulas will measure what you think credibility and behavior is, not what i think, or just common sense.

$133 USD in 2 days

0.0

(0 reviews)

0.0

@singhprashant7

We can analyse your data using a map reduce program. How large your data set is? We can discuss further on this requirement.

$100 USD in 1 day

0.0

(0 reviews)

0.0

@CHRISTIANLA28

I understand the work to extract data because they must organize, tabulate, and prepare a report in Excel with advanced graphics and pivot tables. Also I can do an analysis solver to compare scenarios and take better decisions; in each part I can be prepared a report with graphs and percentages. I have ability to understand each task to realize, I handle excel logical and financial formulas of, as well as to managing and manipulating data for analysis using pivot tables, advanced graphics and scenarios. During my career always manage Excel financial management, at my work I use daily to make projections, formulations, I have over 3 years managing Excel. The first day and the first hours I will be to organize and analyze information. After defining what is the best tool or method which is adapted to manipulate information and to obtain a reliable analysis and comparison of scenarios. Finally graph analysis with advanced graphics, do tables in solver and handle pivot tables providing an organized report page by page. CONNECT MORE PARNERTS JOBS I hope we keep in contact for future work, thank you for giving me the opportunity to show my skills and responsibility.

$100 USD in 3 days

0.0

(0 reviews)

0.0

@aks10000

I must tell you that how it is going to be done. First the relationships between dependent variable and independent variables through Q-Q plot and interactions of independent variables will be analyzed. After understanding the relationships through Q-Q plots, correlations among the variables will be calculated to explore the significant correlations. After this the levels of various independent variables has to be defined to make index i.e. Gearcash Basic Credit Score and Gearcash Advanced Credit Score. The score indices can be back-tested with historical data for improving the model.

$122 USD in 1 day

0.0

(0 reviews)

0.0

@nisargpathakphd

As the professional working in the financial data analytics, I have managed few projects on credit cards domain I have enough domain experience to handle the project. As per the project description it can be easily solved by naive base classification. The result can be refined by the use of pure posterior probability. As a result I will provide you 3 formulas with its proper explanation to justify the result.

$30 USD in 1 day

0.0

(0 reviews)

0.0

@kkdatamanthan

I am assuming you will provide the historical data. Based on historical data, I will start with simple perceptron model and then move to complex model. Complex models are not always better but I will check to see if there is any improvement. I will divide the data in test and training group to see how well my algorithm performs on new data. I have done similar kind of work in medical statistics where I assigned a susceptibility ranking to a person for a particular disease based on patients attribute. In present case, borrower is patient. When developing that algorithm, I derived inspiration from the way credit scores are calculated. Thi project is reverse of that.

$116 USD in 3 days