A year and half into Shinzo Abe's attempt to lift Japan's economy from its two decades' stagnation, policy makers say Abenomics is on track: inflation has moved higher, corporate profits are up and businesses are starting to invest again, while not everyone is optimistic, the Nikkei is the worst performing major  market in the world as investors worry about slow progress on structural reform and looming debt crisis. 

Abe administration definitely is not happy about the latest GDP publish , as Japan GDP fell 6.8 % in Q2 this year majorly due to that the economy is hit by the hike of sales tax. The reversal of GDP is Q2 is within expectation as shoppes would shop crazy in Q1 pulling forward demand before tax increased and leaving a spending plunge in Q2 , but what surprised me , was later I found out most economists predicted  an exact 6.8% decline, and their crystal ball is the " Nowcast" Model.

You might heard of Forecasting and wonder what Nowcasting Model is? At first glance the term" nowcasting" is just a contraction of "now" and "forecasting", and actually Nocasting 's meaning is exact as the two words combination : forecast of now or present.

Let me give an example of Nowcasting's application. We know that standard measures used to assess a state's economy,  eg., gross domestic production (GDP), can only be determined after a long delay and sometimes even are subject to a sebsequent revisions, while weather forecasters only knows the weather conditions today and needs to predict the weather tomorrow.  The prediction of GDP is an application of  "Forecast" model , and the prediction of Weather is of "Nowcast", and if economists just forecast the present or even the recent past, the short term forecast is called " Nowcasting".

Nowcasting model historically have been based on simple heuristic approach, and with the increasing availability of more data in modern era, it has been shown that Nocasting model can be formalized in a statistical model without the need of informal judgement. The model in the paper below exploits information from a large quantity of data sets at different frequencies and with different publication lags. The idea is the signal of GDP change can be extracted from this large and heterogeneous set of information source before GDP published, like jobless figures, the trade balance, industrial orders,etc.

The wide range use of social media and the easy to query content of these media have given Nowcasting the opportunity to be used outside Economics field. Nowcasting methods based on twitter content have been developed to estimate hidden quantities such as the "mood of nation" .

The above is is just a brief introduction of the definition, use and achievement of  Nowcast. If you have more interest of exploring the use of Nowcast, and don't mind read dull and dry paper, one more recommended for you here .