Econometric modelling and quantitative analysis

Start adding items to your reading lists:
or
Save this item to:
This item has been saved to your reading list.

Data, analytics, inference – what does it all mean?

In today’s world of Big Data, businesses and policymakers have an increasing amount of information at their fingertips. Having all of this data to hand, though, is only as useful as the ability to interpret what it means accurately. Indeed, conducting too simple an analysis or using complicated techniques in the wrong way can lead to incorrect inference and bad businesses decisions.

Econometrics can help you to understand what data really tells you about your business, your customers, and the wider market, by being:

  • transparent – as a technique, regressions are not subject to the “black box” critique;
  • flexible – there are a battery of different techniques to suit what output is required; and
  • robust – econometrics is underwritten by an established wealth of academic literature.

We have a team of highly skilled econometricians who work across all sectors, employing a 6-step project-bespoke approach to deliver real insight.

 

So what is Econometrics?

Econometrics is the economic application of regression analysis where historical data is used to analyse the relationships between pairs (or sometimes panels) of different variables.

To understand the relationship between two variables, say X and Y we would need to build a model based on the equation set out in this diagram. This shows that a change in X has an effect on Y to the magnitude of the coefficient beta. However, there are other factors which also affect Y that cannot be accounted for by movements in X – these are captured in the error term, e.

The job of econometrics is to estimate the value of beta based on historical data.  In addition, further variables can be included in the model to 'unpack' other factors which impact the variable of interest (Y).

 

How can we help?

Elasticity analysis

How do price changes affect sales volumes?

Elasticity analysis is used to understand how responsive an economic variable is to another. It is commonly used to understand how sales vary as price changes. However, it can also be applied in other settings, such as estimating how changes in connectivity affect productivity.

View more

Economic forecasting

Forecasts can be carried out at a high level of aggregation—for example for GDP, inflation, unemployment or the fiscal deficit—or at a more disaggregated level, for specific sectors of the economy or even specific firms.

View more

Disputes

Using econometrics to inform competition responses and investigations

We have worked for clients in response to competition investigations and cases in the UK and abroad where we have utilised econometric techniques to construct counterfactual scenarios and analyse anti-competitive activity. 

View more

Impact assessment

Placing a value on how your business impacts the economy, the environment, and wider society is as important as calculating financial return. Altogether, this measures the total impact your business has on society and provides the information you need to navigate today’s operating environment. We believe a total impact approach to making strategic decisions provides the holistic perspective a business needs to understand risk, identify opportunities, and optimise its contribution to society. 

View more

Efficiency testing

What are the barriers to business efficiency?

Stochastic Frontier Analysis can be used to identify the optimal input mix, whether it is a manufacturing company, hospital or university. 

 

View more

Strategy evaluation

Formative evaluation helps shape the project. It includes activities such as gathering information on the audience that might help determine the best way to present information. Formative evaluation begins early in the development of the project and may be collected in phases as the project develops. By conducting economic evaluations of for example health services and interventions through costing exercises and analyses of cost-effectiveness.

 

View more

 

Our bespoke approach

number one

Define the problem

  • What problem are you trying to solve?
  • What do you want the model to be used for?
number two

Collect the data

  • What data are required to solve the problem?
  • Are there novel ways to estimate the data you don’t have? 
number one

Develop the model

  • What outputs are needed?
  • Which statistical problems need to be controlled for?
number one

Test the relationship

  • Are relationships really significant?
  • Does the relationship satisfy hypotheses or statistical tests?
number one

Interpret the findings

  • What is the model telling us? Does this make sense?
  • How can we communicate our results in the most impactful way?
number one

Test for robustness

  • What are the model sensitive to?
  • What are the views of our associated academics? Is further quality assurance required?

Case studies

How do changes in long distance connectivity affect productivity?

We have recently worked for Highways England in building the economic case for a major strategic roads project. As part of this we were asked estimate the causal link between long distance connectivity and productivity.

How do you measure the connectivity of a region?

We used a novel dataset of historical road and air travel times to compute optimal travel times between regions of Western Europe. We then used these to weight a measure of economic mass to determine how connected a particular region is (this is shown on the map below).

How do you estimate the relationship?

We then used advanced econometric techniques to control for unobservable variables that effect connectivity to analyse how a changes in connectivity affect firm productivity at the  firm level.

Our work has been praised by the client and has received exemplary feedback from our academic peer reviewers.

How will a tax on sugar-sweetened beverages affect sales, the industry and the economy?

Our client, one of Colombia’s largest soft drinks manufacturers, commissioned us to help estimate the impact of a tax on sugar-sweetened beverages (SSBs).

How will a sugar-sweetened beverage tax affect sales?

We analysed historic consumer reactions to changes in prices using cutting-edge econometric techniques. The graph below shows how demand responds to changes following a one-time change in price. For example, it predicts that a tax that increased price 10% would decrease demand by 20% after 10 months.

How will reduced sales of SSBs impact the soft drinks industry and the economy?

We used an Input-Output analysis to model the interlinkages between different sectors in the economy. This allowed us to produce reliable estimates for the impact of the tax on economic output, measured in Gross Value Added (GVA), and employment.

The client used out analysis to formulate a business plan and engage with the government. 

Contact us

Jonathan Gillham
Director of Econometrics and Economic Modelling, PwC United Kingdom
Tel: +44 (0)20 7804 1902
Email

Follow us