China Live Economic Forecast

China Live Economic Forecast of GDP Growth Rate, Unemployment Rate, and Inflation rate with the Google Domestic Trends Estimation.

Google Domestic Trends track Google traffic across specific sectors of the economy. Changes in the search volume of a given sector on google provide unique economic insight. 

The indexes measure relative query volume compared to the total number of searches on google.Thus, an index that is decreasing does not imply that the total number of searches is decreasing; in fact the number of queries could be increasing, just not as quickly as overall search volume.

*Google, as an search engine is not available in China for now, the only way to access google in China is to have an VPN. 

*Google has not been completely blocked in China. You can still have access to google map, google translation and few other products. So we use limited data set for China.

OverviewActualGoogle Trend EstimationQ4/17 EstimationQ1/18 EstimationQ2/18 Estimation2020 Estimation
GDP Growth Rate1.701.821.
Unemployment Rate3.954.
Inflation Rate1.401.461.822.43.7
Interest Rate4.354.414.454.394.25.5

Take a look at how we estimate with google data:

Regression Analysis Guide

Estimation with Multi Variables

Can Google queries help predict economic activity?

Interesting Examples
From the graph below you will see that actual clothing store sales and the Google Retail index has been decreasing over time. There are seasonal spikes in clothing store sales, but over time there has been a pronounced decrease.

Google Auto Buyers Index
User searches related to car buying increased dramatically with the launch of the US Government's Cash for Clunkers program. We see that the Auto Buyers Index jumped to 1.4 at the end of July 2009.

Predicting retail sales
The US Census Bureau releases the Advance Monthly Retail Sales survey 1–2 weeks after the close of each month. These figures are based on a mail survey from a number of retail establishments and are thought to be useful leading indicators of macroeconomic performance. Learn more about the survey and the procedures followed in constructing these numbers.

The retail sales data is organized according to the NAICS retail trade categories. The data is reported in both seasonally adjusted and unadjusted form; for the analysis in this section, we use only the unadjusted data for the NAICS category 448 (clothing and clothing accessories stores).

Google Domestic Trends data

The Google RETAIL index tracks query volumes on related to the retail trade category. Sample queries are "retail, bags, retail sales". Thus, to predict retail sales of March, 2008 we use the index value on March 15th, 2008.

It is observed that the Mean Absolute Error improves from 6.52% to 3.65% (an improvement of 44%)!

Methodology for Predictions

Prediction models use a simple linear model:
Base prediction model for Retail: use last month's value and a year back (i.e. )—t refers to current month, t-1 is last month, t-12 is one year earlier.
The index was added as an additional independent variable to the respective base prediction model

Rolling 24-month regression (least squares fit) to get the weights of the terms in the equations using data from 2006-2007. Then using the weights computed predicted values for 2008 and then determine the mean absolute error.

The index value used for predicting was either the 30 day moving average of the index OR the Year-over-Year change in the 30 day moving average.