
Binarium
The Best Binary Options Broker 2020!
Perfect Choice For Beginners!
Free Demo Account!
Free Trading Education!
Get Your SingUp Bonus Now!
CFO Forecasting – Financial Forecast Sandesh Brand 3 – Refinement
Key Information
Challenge Overview
Background
From 25th January 2020, a challenge was run to generate a series of high quality financial forecasts for a consumer broadband brand. This challenge was called ‘CFO Forecasting – Financial Forecast Sandesh Brand 3’. It generated strong results on 3 of the 6 target variables across the two target products while the other 3 variables require further work.
This challenge is designed to improve on the accuracy and precision of these 3 lower performing forecasts to reduce MAPE and individual APE results.
The input data sets have been Normalised since the previous challenge as outlined below to support refinement.
The leading models from the initial challenge in January are shared as a possible foundation for refinement though these need not necessarily be used if superior results can be achieved though alternative approaches.
Challenge Objective
The objective of this challenge is to generate the highest accuracy predictions possible for the 3 financial variables outlined below, for each of the two products.The accuracy of the forecast must at least improve on the Threshold target quoted for each variable / product.
The model should be tailored to a 12mth forecast horizon but must be extendable beyond this time period.
The accuracy of a prediction will be evaluated using MAPE (Mean Absolute Percentage Error) and maximum APE (Absolute Percentage Error) on the privatised data set over a period of 7 – 9 months.
Business context
The two products are broadband products:
Tortoise (legacy product, declining product, available everywhere) – slow download speeds.
Falcon (main product, reaching maturity, available in most of the country) – faster download speeds.
These two products do have an inter dependency since Falcon product is an upgrade of the earlier version, with the product growth of the later version dependent to a large extent on upgrading the customers from the earlier version. There is therefore a gradual move from Tortoise to Falcon.
The six variables are financial metrics

Binarium
The Best Binary Options Broker 2020!
Perfect Choice For Beginners!
Free Demo Account!
Free Trading Education!
Get Your SingUp Bonus Now!
Gross adds – the number of new subscribers by product joining the brand during a month
Churn or Leavers – the number of subscribers by product who terminated service with the brand during that month
Net migrations – the number of subscribers who remained with the brand but moved to another product. This usually is an upgrade to faster broadband speed.
These three ‘discontinuous’ variables are seasonal – see Business insight section for more detail; can vary significantly from month to month; and are almost entirely dependent on the market, and competitor pressures at that point in time.
Challenge Thresholds and Targets
Models 348636 and 349512 are provided as foundation. These models can be used to baseline current performance on Privatised data set from which improvement will need to be made.
Note: Performance on Privatised data set may not correlate directly with performance on Real data set.
Your submission will be judged on two criteria.
Minimizing error (MAPE).
Achieving the Thresholds and Targets designated in the tables above.
The details will be outlined in the Quantitative Scoring section below.
Business Insights
The relationship between key financial variables
Closing Base for a Product = Volume Opening Base for that Product + Gross Adds – Leavers + Net Migrations to that Product
Net Migrations
Net migration is the difference in the number of existing customers with Sandesh Brand 3 that move onto and off a specific broadband product per month.
Net Migrations for Tortoise are a direct mirror image of Net Migrations for Falcon such that Net Migrations for Tortoise + Net Migrations for Falcon = 0 for each month.
This means that it should be possible to forecast both variables with equal accuracy and precision.
Trading Seasonality due to Accounting policies
These three variables are monthly trading performance figures and are calculated based on ‘trading months’ rather than ‘calendar months’. Each ‘trading month’ is artificially constrained to either 4 weeks or 5 weeks exactly, and are arranged in a 4 week, 4 week, 5 week pattern every 3 months. This artificial construct distorts the performance figures.
In the privatised data sets provided, the monthly performance has been Normalised to a consistent, standardised 4.3 week month. This approach significantly smooths the time series data set as demonstrated in the attached diagrams.
This is particularly true from August 2020. Prior to this date, the ‘trading seasonality’ is not so clear cut but the data set has been treated consistently throughout.
It is anticipated that this Normalisation of the variables will make it easier to forecast their performance and improve accuracy and precision.
Gross Adds for both Products – Real data set prior to Privatisation
Churn (Leaver) for both Products – Real data set prior to Privatisation
Net Migrations for both products – Real data set prior to privatisation
Note: Tortoise and Falcon are mirror images of each other
Churn / Leavers: Price increases impact and other external factors
Churn performance has been directly impacted by two external factors. It is assumed that building these appropriately into the Churn forecast, will improve forecasting performance. See diagram below for details.
These external factors are two fold
Regular Price Increases to the existing Customer base.
These increase have occurred once a year though exact timing varies year to year
The size of the price increase has however been the same every year
The impact of the price increase predates the price rise (by 1 or 2 months) and extends for up to 5 months afterwards as customers react to the increase.
The impact of these regular increases needs to be factored into the model
Any treatment needs to be clearly documented and reproducible on the real data set.
A single, oneoff ‘shock’ in Sept 2020.
This was a ‘oneoff’ shock and will not be repeated in the future
The size of the impact has not been quantified – part of the challenge
The length of the impact is known to be from Sept 2020 to Dec 2020.
This ‘oneoff’ impact overlaps with the price increase in 2020
The impact of this oneoff shock needs to be estimated and treated so as not to impact future predictions.
Any treatment needs to be clearly documented and reproducible on the real data set.
Price Increase and ‘oneoff’ impacts – see table below
Financial Year modeling:
Sandesh reports its financial year from April – March. This may contribute to seasonality based on financial year, and quarters (Jun, Sep, Dec, and Mar), rather than calendar year.
Anonymised and Privatised data set:
‘Zscore’ is used to privatise the real data.
For all the variables, following is the formula used to privatise the data:
where z i = zscore of the ith value for the given variable
x i = actual value
μ = mean of the given variable
σ = standard deviation for the given variable
Modeling Insight derived from previous challenges.
Optimise the algorithms by minimising RSME
It is recommended to optimise the models by minimising RSME, rather than MAPE because of the privatisation method used. It is strongly believed that minimising RSME will create the best model capable of being retrained on the real data set.
Final Submission Guidelines
Submission Format
You submission must include the following items
You are asked to provide forecasts for the following 12 months based on the given dataset. We will evaluate the results quantitatively (See below). The output file should be a CSV format file. Please use the “Generic LookupKey” values of the target variables and “Date” as the header.
A report about your model, including data analysis, model details, local cross validation results, and variable importance.
A deployment instructions about how to install required libs and how to run.
Expected in Submission
Working Python code which works on the different sets of data in the same format
Report with clear explanation of all the steps taken to solve the challenge (refer section “Challenge Details”) and on how to run the code
No hardcoding (e.g., column names, possible values of each column, . ) in the code is allowed. We will run the code on some different datasets
All models in one code with clear inline comments
Flexibility to extend the code to forecast for additional months
Quantitative Scoring
Given two values, one ground truth value ( gt ) and one predicted value ( pred ), we define the relative error as:
MAPE( gt , pred ) =  gt – pred  / gt
We then compute the raw_score(gt, pred) as
That is, if the relative error exceeds 100%, you will receive a zero score in this case.
The final MAPE score for each variable is computed based on the average of raw_score , and then multiplied by 100.
Final score = 100 * average( raw_score(gt, pred) )
MAPE scores will be 50% of the total scoring.
You will also receive a score between 0 and 1 for all the thresholds and targets that you achieve. Each threshold will be worth 0.033 points and each target will be worth 0.05 points. Obviously if you achieve the target for a particular variable you’ll get the threshold points as well so you’ll receive 0.083 points for that variable. Your points for all the variables will be added together.
Judging Criteria
Your solution will be evaluated in a hybrid of quantitative and qualitative way.
We will evaluate your forecasts by comparing it to the ground truth data. Please check the “Quantitative Scoring” section for details.
The smaller MAPE the better.
Please review the targets and thresholds above as these will be included in the scoring.
The model is clearly described, with reasonable justifications about the choice.
The results must be reproducible. We understand that there might be some randomness for ML models, but please try your best to keep the results the same or at least similar across different runs.
Payments
Topcoder will compensate members in accordance with our standard payment policies, unless otherwise specified in this challenge. For information on payment policies, setting up your profile to receive payments, and general payment questions, please refer to https://help.topcoder.com/hc/enus/articles/217482038PaymentPoliciesandInstructions
Reliability Rating and Bonus
For challenges that have a reliability bonus, the bonus depends on the reliability rating at the moment of registration for that project. A participant with no previous projects is considered to have no reliability rating, and therefore gets no bonus. Reliability bonus does not apply to Digital Run winnings. Since reliability rating is based on the past 15 projects, it can only have 15 discrete values.
Read more.
Master List of All Forecasts Produced by FFC
Below is a list of all forecasts produced by the Financial Forecast Center. The first link takes you to the freely available forecast. The second link takes you to the long range forecast which requires a subscription to access.
Checking your browser before accessing enotes.com.
This process is automatic. Your browser will redirect to your requested content shortly.
How Financial Modeling Differs from Financial Forecasting
June 12 2020 Written By: EduPristine
I have observed often that students (and even professionals) get confused with Financial Modelling and Financial Forecasting. These two terms are often used interchangeably as well (mostly in a wrong context). These are neither exactly the same thing nor they are two disjoint sets. Let’s try to understand these concepts with the help of an example.
Imagine that you are working as a Credit Risk Officer in a bank. The Rolling Motors, a car manufacturing company, is one of the bank’s existing corporate clients. The company already has taken two loan facilities from the bank with which they have setup two car manufacturing units, one in Pune and another one in Chennai. Now the company wants to setup another car manufacturing unit in Thane, Mumbai and has approached the bank for a third loan facility. You, as the credit officer, will have to evaluate this proposal and take a decision if the bank wants to extend the third loan facility to the company.
Basically, there would be two parts you would have to evaluate. In the first part [part1], you would like to understand how well the Rolling Motors Company has been performing till now and their conduct of the existing two loan accounts. In the second part [part2], you would try to assess if the proposed third plant would make sense for the company to setup and if it would make sense for the bank to extend the third loan facility.
Historical Financial Modelling:
For part1, you need to take their historical financial numbers and put in a spreadsheet in a structured format. You would like to understand the company’s sales growth YoY, improvement in gross profit margin over last few years, the major cost heads such as raw material, labour cost, plant maintenance cost and if these cost items as a percentage of revenue have remain within an acceptable level for last few years, EBITDA margin, net cash flow, current ratio movement, workingcapital requirement, debt servicing etc. This part is known as the Historical Financial Modelling which is used to assess a company’s current state of affairs and how it has performed over last few years.
Financial Forecasting:
As regard to part2, you’ll have to understand the future benefit to the company by settingup the third manufacturing plant. To assess the future benefit, we first need to start with forecasting car demand in the market for next few years to see if there would be consumer appetite in future to buy the type of cars that the Rolling Motors would manufacture from the third plant. Based on this demand forecast, you can estimate the sales the third plant can possibly generate for the company. There would be other things to consider here as well – such as the projected market share of Rolling Motors, the future selling price of a car, any regulatory change that can impact the car sales (eg. BSIV emission standard) etc. When you have a realistic estimate of the revenue, you can gradually forecast the other line items such as raw material cost, labour cost etc. This would help you to prepare the estimated cashflow that the third plant can generate in next few years and if that would be sufficient to meet the debt repayment obligation for the third loan facility. This part is known as the Financial Forecasting Modelling.
Together we have Financial Modelling:
In this context, Financial Modelling primarily comprises of two parts – historical performance analysis and future performance prediction i.e. Forecasting. Therefore, Financial Modelling and Financial Forecasting are not exactly two different things, rather Financial Forecasting is a subset of the holistic Financial Modelling exercise.
Different forecasting methods:
Depending on situations, there could be different types of parameters that one need to forecast in the context of financial modelling, such as demand forecasting, sales forecasting, unit price forecasting, raw material price forecasting etc. We can employ qualitative or quantitative or a combination of both methods as deemed suitable for a case.
Qualitative forecasting methods employs surveys and depends heavily on viewpoints of experts in a particular field. This method is particularly useful to predict the shortandmedium trends, but has limitations of biasness over quantitative method of forecasting.
 Delphi Method – Opinion survey with the domain experts and then compiling them to forecast;
 Consumer Survey to understand the shift in consumer choice before launching a new product or service;
Quantitative forecasting methods uses data from the past, evaluate causal relationships among parameters to predict the future numbers.
 Time series method uses the historical trend in a parameter and weigh that with the recent available data to forecast for future;
 Econometric modelling involves robust mathematical and statistical calculations. This type of forecasting modelling is primarily used in the academic field or a situation that involves a large set of interdependent parameters such as in case of economic policy formulation;
Each of these methods requires separate discussion for detail understanding. I’ll take up each of these methods in my future articles.

Binarium
The Best Binary Options Broker 2020!
Perfect Choice For Beginners!
Free Demo Account!
Free Trading Education!
Get Your SingUp Bonus Now!