Forecasting personal consumption expenditures (PCE) with Indicio

Forecasting personal consumption expenditures (PCE) with Indicio

Macrobond users can now access Indicio’s technology through a direct API.

May 25, 2023
Stavros Malkidis with Arnaud Lieugaut

Will the Federal Reserve truly pause rate increases if inflation keeps running worryingly hot?

The Indicio forecasting tool suggests that Friday’s US data release will show that the PCE measure of inflation rose at a 4.3 percent year-on-year pace. This matches the “nowcast” by the Cleveland Fed, but outpaces the 4.1 percent forecast from Trading Economics. Inflation is then expected to decelerate through May and June, according to Indicio.

Macrobond blog: Forecasting personal consumption expenditures (PCE) with Indicio

<p class="blog-chart-link">Macrobond users, access the chart here</p>

With inflation in focus, we have generated a forecast for US personal consumption expenditures (PCE) with Indicio.

Indicio is an automated, machine-learning platform that allows you to estimate univariate and multivariate time series models to forecast macroeconomic and financial data.

While the consumer price index (CPI) is perhaps better known as an inflation gauge, the Federal Reserve is known to pay close attention to measures of PCE, rather than CPI, which puts a larger weight on shelter, food and energy.

(We have previously written about the real-time differences between CPI and PCE and the implications for monetary policy here.)

Academic research suggests that it can be particularly challenging to forecast short-run inflation dynamics. Indicio aims to combine different modelling approaches, potentially creating a super-forecast that can outperform any single model.

“Traditional” macroeconomic data is released quarterly or monthly at best, and we are only able to observe the actual numbers long after the period referred to. On top of that, this data is frequently revised. 

Therefore, models rely on univariate and multivariate regression techniques that crunch higher-frequency indicators of weekly or even daily series.

Indicio helped us select several monthly and weekly variables by measuring their statistical influence on PCE, which were then used to generate different univariate and multivariate models, including various vector autoregression (VAR) and vector error correction models (VECM).

We retained the following indicators: U.S. CPI, retail trade, U.S. core PPI and retail gasoline prices.

As a final step, Indicio allowed us to generate a single forecast, weighting the various models estimated based on their stepwise root mean square error (RMSE).

April values for our four explanatory variables have already been released. Using these, we generated a scenario forecast in Indicio – effectively creating a “nowcast” of PCE.

In the newest version of Indicio, users can export their outputs back into Macrobond, and use our front-end to chart their models, like we did here.

As the chart shows, Indicio predicts a 4.26 percent increase for PCE in April – a small acceleration year-on-year compared to March.

(We chose to chart this against the estimates from the various relevant VAR / VECM univariate and multivariate models we used to generate the weighted forecast.)

Indicio can allow you to rapidly build a wide range of sophisticated, statistically robust forecasts with zero coding. Read more here.

All written and electronic communication from Macrobond Financial AB is for information or marketing purposes and does not qualify as substantive research.