What is Quantitative Analysis (QA)?

Quantitative analysis is like the way Mr. Spock views the world…

In the classic Star Trek television series, Mr. Spock was a character from the planet Vulcan. He tended to rely heavily on logic and to avoid the biases caused by emotions. In the show, Mr. Spock was portrayed as cold and emotionally void. He only cared about the facts of the matter at hand, approached every decision from the numerical trade-offs, and discounted the sentimental value others attached to things. In other words, he relied almost solely on informal quantitative analysis.

A quantitative analysis relies on data. It is an assessment rooted in the ability to quantify something. Whether it’s money, time, temperature, votes, or the chemical reactions in a solution, the target variable can be measured and recorded as a number.

With that information, analysts can employ statistical methods to gain insight into what all that data means. However, there are just some things in life, and in business, that don’t translate into a spreadsheet very well. Qualitative analysis is when you use your eyes, heart, and gut instinct to capture information that the numbers might not. For example, how compassionate is the chief executive officer? How is the morale of the employees? And what is the board of directors’ stance toward supporting local artists?

Quantitative analysis tells you how things have been, are doing, or will be. It uses a record of performance and a description of the output to make inferences about the future. Evaluating the data can expose trends and changes that might not stand out without seeing the numbers. But the numbers are only as good at the quality of the data. Sometimes people can see a problem before it shows up in the data. And good leadership can use qualitative analysis to head those problems off so that they never do.

Qualitative and quantitative analysis are not opposites of each other. In most circumstances, they act in concert to help you gain strategic insight and complete information. Relying too heavily on either approach while neglecting the other is a good recipe for making bad decisions.

Quantitative analysis (QA) shows up in many fields of study. That includes business management, finance, and economics.

In business, a manager might use QA to forecast sales, plan production, improve profits, and manage inventory. In finance, an analyst might use QA to determine the value of a company, evaluate an investment, and assess the chances of default of a loan. Economists use QA to understand the direction an economy is moving, determine weaknesses in a supply chain, and consider the risk within a portfolio.

There are many types of QA conducted every day in businesses around the world to aid in decision-making.

Some are simple personal finance decisions about budgeting. Others are evaluating the right price in complex corporate mergers. And, with the explosion of quantitative data being captured in our ever-increasingly digital world, there is more and more demand for people who can translate that data into useful insights.

You can think about quantitative methods in five general categories. Each has a different level of sophistication and a different goal.

Traders are often evaluating companies and looking for buying opportunities. They might pore through data about sales, revenues, and costs. Within all of this research, there are several ways to adjust information about very different companies so that they can be directly compared.

For example, an analyst might divide the company earnings by the number of outstanding shares to determine the earnings per share. Or, they may divide the market capitalization of a company by its earnings to derive the P/E ratio. Comparing the EPS, P/E ratio, and other financial ratios of two companies might help the trader decide which one to invest in.

Economists and statisticians regularly use quantitative analysis to identify trends and relationships between data sets. By gaining insight into the direction things are moving, analysts can attempt to predict what will happen next. And by understanding how changes in one variable influence others, they can try to forecast how fluctuations will ripple through a system.

One popular technique for spotting trends is the moving average. And perhaps the most commonly used method for discovering relationships is called regression analysis. Both of these techniques have simple and complex forms, and each has a variety of uses.

But predictive analytics provide insight into how the future should unfold, and how it might unfold differently if alternative decisions are made. For example, a moving average can help a business owner understand the direction sales are heading, and a regression can help them predict how much those sales might change if they increase their marketing budget.

Many quantitative analysts use models to gain a better understanding of an investment, company, or financial instrument. These financial models are scaled-down versions of reality that capture how money moves through a system and how decision points alter outcomes.

For example, a financial model might follow an increase in marketing as it turns into increased sales, which in turn requires increased production, which may require additional sourcing at a higher cost.

These models are vital in project management, as they can simulate how the supply chain, cash flow, labor needs, equipment availability, financing costs, and inventory may change over time. Thus, it can flag problems and pinch points that may constrain the business operations.

Other quantitative models use approaches such as discounted cash flows to determine the value of a company or investment opportunity. In many cases, outputs from predictive analytics serve as inputs into these models.

One of the essential uses of quantitative analysis in business is to optimize operations. For example, analysts can use a technique called linear programming to determine the profit-maximizing or cost-minimizing combinations of production levels.

In other situations, computer software programs can test thousands of options within a few minutes to locate the optimal strategy for anything from which equipment to use to labor allocations to delivery paths.

In the digital age, everything is data. The amount of a sale, the frequency that a shopper visits, even how long they spend in the store can be recorded in a database. These days, business analysts might end up with more information than they know what to do with.

Within those databases are hidden pieces of information that could be valuable to a company. But dealing with massive datasets is challenging. Data mining allows analysts to locate patterns and relationships hidden in the billions of observations. And with that information, businesses can position themselves to be ready to meet their customers’ needs before the customers even realize what they are.

The most popular of these unique types of quantitative analysis tends to focus on classification and pattern recognition.

Classification techniques include cluster analysis, which locates groupings within seemingly random data. By identifying which customers belong to which cluster, the company can improve the effectiveness of its marketing by designing targeted ads to that customer group.

Pattern recognition programs can search datasets for correlations between all sorts of information. While such a shotgun approach to statistics is not a scientifically valid way to form and test hypotheses, data mining can locate hidden patterns between variables that statisticians would not think to hypothesize.

Once found, the company can watch to see if the correlations continue or if they were just a random occurrence in the data. Successfully identifying relationships in the data can give a business an upper hand, as they can better predict what customers are going to be looking for, and can plan accordingly.