Once you identify the most important objective, the next step is to consider the data requirements. What will you need to achieve the required work outcome?
Which data sets should you analyse to meet your objectives?
Needs depend on whether the desired outcome is quantitative (numerical) or qualitative (categorical). Numerical outcomes are quantified, counted, or measured, and given a numerical value, so variables might apply to height, weight, and age measures.
In contrast, categorical outcomes are descriptive and expressed in words rather than numbers. Data can take on one of a limited (usually fixed) number of values, assigning each observation to a particular group. Category variables include race, gender, age group, and educational level.
Different data types appear in the next image.
Watch
Watch the video to clarify the difference between quantitative and qualitative.
Note: Quantitative and qualitative data may be required to investigate one problem statement.
First-party data – like internal customer satisfaction surveys or sales figures – is collected from within the business.
Examples include:
- sales records
- receipts
- invoices
- profit and loss statements
- records of expenditure.
Second-party data – like website or social media information – is collected from other organisations. Third-party data, like census or market research information, is collected from multiple sources.
External financial data is generated from outside the business, including stock market, interest rate, and economic data. This data can be found in economic reports released by the government.
Examples of external financial data include:
- raw material costs
- mortgage records
- credit rating assessments
- customer profiles.
Determine data requirements by considering data types and their suitability to achieve the desired work outcome.
Several statistical tools and techniques are used in the financial services industry.
Statistical tools can be used to describe current circumstances, identify patterns and trends, assess financial performance, determine relationships between different aspects of a business, inform decision-making, and present data to stakeholders.
In this module, statistical tools and techniques include but are not limited to financial ratios and statistical modelling.
Financial ratios
Ratios are used to compare two or more sets of data.
They reflect:
- the percentage difference, which is used to compare two percentages and find out how much they differ from each other
- the percentage ratio, which compares two percentages and finds out what percentage of one number is equal to the other number
- the proportion, which is used to compare two sets of data and find out if they are in proportion to each other
- the correlation coefficient, which measures the strength of a linear relationship between two sets of data.
In mathematical equations, a ratio indicates how many times one number contains another. For example, if there are eight oranges and six lemons in a fruit bowl, the ratio of oranges to lemons is eight to six. Similarly, the ratio of lemons to oranges is 6:8, and the ratio of oranges to the total amount of fruit is 8:14.
Liquidity ratios
There are several liquidity ratios that analysts use to assess financial health. The most common ratio is the current ratio, which measures a company's ability to pay its short-term obligations with its current assets. A high current ratio indicates that a business has enough liquid assets to cover its short-term debts.
Another important liquidity ratio is the quick ratio, also known as the acid-test ratio. This measures a company's ability to pay its short-term obligations with its quick assets, which are liquid assets that can be quickly converted into cash. A high acid test ratio indicates that a company has plenty of liquid assets to cover its short-term debts.
To calculate the liquidity ratio, divide current assets by current liabilities. This will give you a ratio that measures how well the business can pay off its short-term obligations. Alternatively, divide cash and assets that can be quickly converted into cash by the current liabilities. This calculation method is a bit more conservative, and some analysts prefer it because it gives a clearer picture of true liquid assets.
Note: This ratio type is good for answering questions about current circumstances.
Current ratio: Current assets ÷ current liabilities
Quick ratio: Cash and equivalents + accounts receivables ÷ current liabilities
Profitability ratios
Profitability ratios are useful to determine how much profit a company has made in relation to its revenue, expenses, and other factors so investors can get a clear picture of its overall financial health.
There are several profitability ratios, each of which measures different aspects of a company's business. The most important thing to remember is that these ratios should be considered in conjunction with each other rather than in isolation. By considering all profitability ratios, you can comprehensively view a company's financial strength.
Some examples and definitions appear next.
ROA measures how profitable a company is relative to its total assets. The higher the ROA, the more efficient a company uses its assets to generate profit. To calculate the ROA, divide the net income by the total assets. This will give you a percentage representing how much profit the business makes for every dollar it has in assets. A high return on assets indicates that the business is profitable and efficient. A low return on assets may indicate that the company needs to improve its operations. To get a more accurate picture of a company's profitability, compare its return on assets to other companies in the same industry. This will help you see whether the business performs better or worse than its competitors.
ROE measures how profitable a company is relative to its shareholders' equity. The higher the ROE, the more profit a company generates for each dollar of shareholders' equity. To calculate the ROE, divide the net income by the shareholders' equity. The net income is the total earnings after taxes, and the shareholders' equity is the total value of the outstanding shares. If a business has a net income of $10 million and shareholders' equity of $50 million, its return on equity would be 20%. A higher return on equity indicates that the business is more profitable and efficient in using its investment to generate profits.
The ROI ratio measures how much profit a company generates for every dollar invested. It is a key metric for assessing a company's financial health and performance. To calculate the ROI, take the profit from spending and divide it by the amount of money invested. This will give you a percentage representing how much profit you made for every dollar invested. For example, let's say the business was considering investing $100,000 in a new product line. After one year, the product line has already generated $120,000 in revenue. To calculate the ROI, divide the $20,000 in profit ($120,000 in revenue - $100,000 in investment) by the $100,000 investment. This gives you an ROI of 20%.
The profit margin measures the percentage of revenue that a business keeps as profit after accounting for all expenses. The higher the profit margin, the more efficient the business generates profits. To calculate a profit margin, subtract your total revenue from sales and your total costs of goods sold. This will give you your gross profit. Then, divide your gross profit by your total revenue to get your margin percentage. If you had total revenue of $100 and your total costs of goods sold were $60, then your gross profit would be $40. To get your margin percentage, divide $40 by $100 to get 40%. In contrast, the assets turnover ratio measures how well a business uses its assets to generate sales. A high assets turnover ratio indicates that a business uses its assets efficiently to generate sales. A low asset turnover ratio may indicate that a company is not using its assets efficiently and could be improved. To calculate the assets turnover ratio, divide the business sales by its total assets. This will give you the number of times assets were used to generate sales.
The gross margin ratio is a profitability ratio that measures a business' manufacturing and distribution efficiency during the production process. The gross margin ratio is calculated by dividing gross profit by revenue. The gross margin ratio is relevant to the analysis of financial data because it provides insight into a company's ability to generate profit from its operations. Calculate the gross margin ratio by taking the total revenue and subtracting the cost of goods sold, then dividing that number by the total revenue. The resulting number will be the company's gross margin percentage. For example, if a business had total revenue of $100,000 and their cost of goods sold was $40,000, their gross margin percentage would be 60%.
The net profit margin measures the percentage of total revenue that a company keeps as profit after accounting for all expenses. This ratio is important because it provides insight into a company's ability to generate income and its overall profitability. Investors often use profitability ratios to compare companies within the same industry. By doing so, they can better understand which companies are more efficient and generate more profits relative to their peers. Additionally, analysts may use profitability ratios to forecast future earnings and identify trends over time. To calculate a business' net profit margin ratio, you will need to know its total revenue and total expenses. Total revenue is the amount of money a business brings in from its sales. Total expenses include the cost of goods sold, operating expenses, and taxes. The formula for calculating the net profit margin ratio is (total revenue - total expenses) ÷ total revenue. This ratio can be expressed as a percentage or as a decimal. A high net profit margin indicates that a company is very profitable and has a lot of room to grow.
The average stockholders' equity ratio is a profitability ratio that measures the average amount of equity held by stockholders over a certain period. The purpose of this ratio is to gauge how well a company is performing and provide relevant information for the analysis of financial data. To calculate an average stakeholders' equity ratio, take the total amount of assets, subtract the total liabilities, and divide this number by the total number of outstanding shares.
Note: This ratio type is good for answering questions about past performance. Analysts also use historical data to predict future performance.
Read
A useful industry resource to understand and calculate profitability ratios is available here.
Activity analysis ratios
Activity analysis ratios measure business efficiency and productivity.
A few different activity analysis ratios can be used, and each one measures something different. The most common activity analysis ratios are the inventory turnover ratio, the accounts receivable turnover ratio, and the fixed assets turnover ratio.
The inventory turnover ratio measures how quickly a company can sell its inventory. The accounts receivable turnover ratio measures how quickly a company can collect payments from its customers. And finally, the fixed assets turnover ratio measures how efficiently a company uses its fixed assets.
Activity analysis ratios can be very useful in assessing the overall health of a company. By looking at these ratios, companies can identify areas where they need to improve.
Note: This ratio type is good for answering questions about efficiency and productivity.
Inventory turnover ratio: Cost of goods sold ÷ the average inventory for the same period
Accounts receivable turnover ratio: Annual credit sales figure ÷ by the amount of money currently owed
Fixed assets turnover ratio: Net sales or revenue ÷ by the average total assets
A useful resource to understand and calculate activity analysis ratios is available here.
Capital structure analysis ratios
Two main ratios are used when conducting a capital structure analysis – the debt ratio and the interest coverage ratio.
The debt ratio is a measure of a business' financial leverage. To calculate the debt ratio, divide the total liabilities by shareholder equity. A higher debt-to-equity ratio indicates that a company is less likely to get funding and has more risk.
The interest coverage ratio measures a company's ability to pay its interest expenses. This ratio is calculated by dividing a business' Earnings Before Interest and Taxes (EBIT) by its interest expense. A higher interest coverage ratio indicates that a company is more able to pay its interest expenses and is therefore less risky.
Note: This ratio type is good for answering questions about borrowing ability.
Debt ratio: Total debt ÷ total assets
Interest coverage ratio: Earnings before interest and taxes (EBIT) ÷ by the total amount of interest expense on all outstanding debts
Read
Read the article to learn more.
Capital market analysis ratios
When analysing a company's stock, there are four main ratios to consider.
They are:
- The price-earnings (PE) ratio. The PE ratio measures how much investors are willing to pay for each dollar of a company's earnings. A high PE ratio means investors are willing to pay more for the company's earnings and vice versa.
- The market-to-book ratio. The market-to-book ratio measures how much the stock market values a company's assets. A high market-to-book ratio means that the stock market values a company's assets highly and vice versa.
- The dividend yield ratio. The dividend yield ratio measures how much a company's earnings are paid out in dividends. A high dividend yield means that more of a company's earnings are paid out in dividends, and vice versa.
- The dividend payout ratio. The dividend payout ratio is a company's annual dividend payments divided by its earnings. This metric determines how much of a company's earnings are paid out in dividends to shareholders. The payout ratio is important because it can give insight into a company's ability to continue paying dividends. If a company has a high payout ratio, it may be using most of its earnings to pay shareholders, which could put future dividend payments at risk if earnings decline. On the other hand, a company with a low payout ratio may have more room to increase its dividend payments in the future if earnings growth continues. For this reason, investors often look at the dividend payout ratio when considering whether to invest in a particular stock.
Note: This ratio type is good for answering questions about value.
Price earnings ratio: Share price ÷ earnings per share
Market to book ratio: Current closing price of the stock by the most current quarter's book value per share
Dividend yield ratio: Dividend per share ÷ market value per share
Dividend payout ratio: Yearly dividend per share divided by the earnings per share
Statistical modelling
A significant portion of analysts focus on statistical modelling to complete predictive analyses.
Watch
Watch the video to learn more about statistical modelling.
Read
Review additional information about statistical modelling here.
Common statistical modelling techniques are included next.
Regression modelling
Data analysts use regression models to examine relationships between variables and determine which independent variables hold the most influence over dependent variables. To clarify, a business might test whether the number of units sold is an appropriate cost driver to estimate marketing costs. The result of this test would confirm whether there was a strong relationship between the number of units sold and the cost of the marketing department. If there was a strong relationship, the number of units would be used to help forecast costs, and if there wasn't a strong relationship, another cost driver would need to be selected and tested.
Linear regression uses a linear equation to model the relationship between two variables, where one variable is dependent, and the other is independent. If one independent variable is used to predict a dependent variable, it is called simple linear regression. If more than one independent variable is used to predict a dependent variable, it's called a multiple linear regression.
Watch
Watch the video to learn more about regression modelling.
The simple logistic regression model can predict the probability that one of two events will occur depending on one variable.
Read
Additional information about simple regression is available here.
The multiple logistic regression model can be used to analyse the relationship between a single event and several independent variables.
The graph below shows how the probability of passing an exam changes depending on the amount of study.
Multiple logistic regression is more accurate than the simple (two-point) logistic method as it considers many more data points, thereby increasing the accuracy of the output.
The discriminative analysis model can classify observations into non-overlapping groups based on scores on one or more quantitative predictor variables. For example, a doctor could perform a discriminant analysis to identify patients at high or low risk for stroke.
Analysis of variance model ANOVA
The ANOVA or high-low method is a data analysis technique that separates fixed and variable costs in a limited data set. It involves taking the highest and lowest levels of activity and comparing the total costs at each level.
Read
Further information about the high-low method is available here.
Note: The high-low method and regression analysis could be used to understand how certain factors have affected performance and forecast future sales, but regression analysis may be more accurate.
Sampling techniques
Sampling techniques are used to select an individual group from a population to study the whole population.
Probability sampling is the most popular method of sampling and consists of selecting a unit from a population such that each unit has a known probability of being selected. This type of sampling is often used in opinion polls and market research.
Non-probability sampling is a method of sampling where units are selected in a way that does not give all units in the population an equal chance of being selected. This type of sampling is often used in convenience samples or when obtaining a list of all units in the population is difficult.
An example of a sampling technique is random sampling. This is where each member of the population has an equal chance of being selected for the survey. This technique can be used to select a representative sample from the population, which can then be used to make inferences about the whole population.
To interpret survey results, group responses into similar categories and count the most common consensus.
Read
More about sampling techniques is available here.
Every statistical test has assumptions that must be met to determine whether the test can be used. If, for example, you wanted to use a t-test, but your data failed the assumptions for this test, you may have to use a Mann-Whitney U test instead.
A few of the most common statistical assumptions are normality, linearity, and equality of variance.
Read
Additional information about statistical assumptions is available here.
Develop a strategy to collect appropriate data depending on the desired work outcome.
Organisational policies and procedures may define what to analyse in each scenario.
If not, outline a strategy that identifies:
- the desired work outcomes
- data requirements
- data types (first, second and/or third party)
- the source
An example appears next.
| Desired work outcomes | Data requirements | Data types | Sources |
|---|---|---|---|
| Produce a monthly sales report – descriptive analysis | Sales ($) by region Sales ($) by product Number of sales Average customer sale ($) |
First party, quantitative | Profit and loss statement Cash flow report Customer Management System (CMS) |
Consider five key questions to select an appropriate statistical method.
For example:
- How many variables do you have to consider? A variable is any characteristic, number, or quantity that can be measured or counted. Variables are sometimes called qualitative or quantitative data items.
- What is your statistical objective? A statistical objective or desired work outcome is usually to describe or classify something, diagnose or explain something, predict or compare something or prescribe suitable actions.
- Determine the scale of measurement used for the variables. Scales are either numerical (quantitative) or categorical (qualitative).
- Determine whether the variables are independent or dependent. Independent variables are the cause, so their value is independent of other variables in your study. In contrast, dependent variables are the effect, so their value depends on changes in the independent variable. Review the examples to learn more.
- Are the variables time or location specific?
A flow chart to identify appropriate statistical tests appears next.
Source: Stackexchange.com
Watch
Watch the video to learn more about statistical tools and techniques.
Consider who might have generated or stored the required data to find what you need.
Financial information sources include:
- Reserve Bank of Australia
- Australian Taxation Office (ATO)
- Australian Bureau of Statistics (ABS)
- National Statistical Service (NSS)
- Australian Securities and Investments Commission (ASIC)
- Australian Prudential Regulation Authority (APRA)
- industry association data and information
- organisational data and information (public record)
- international government and organisation data and information.
Access first-party data by reviewing financial records.
Seek feedback
If the goals are unclear or additional information is required, ask questions in a face-to-face meeting via email or telephone.
For example:
- What data do you want me to interpret and analyse – why?
- What do you think the data shows – why?
- Is this the only data you have for this time period?
- Who provided the data?
- How was the data generated?
- When was it generated?
- How do you want me to collect additional information?
- How will the data be processed?
- What type of report do you want?
If you must, make decisions yourself in consideration of the legal and ethical requirements. Statistical analysis must be accurate and consistently aligned with industry and organisational standards.
To make high-level decisions autonomously, review the organisational goals and ask the supervising manager what they want to achieve from the analysis. Good decision-making is based on research and discussion wherever possible.
Follow the steps below to ensure your analysis helps to achieve and improve organisational goals:
- identify and understand the issue at hand
- analyse the extent of the issue, and determine what urgency there is for the issue to be addressed
- determine the best way to respond through consultation.
