What Is the Retail Price Index?

The retail price index, compiled by the National Bureau of Statistics, is an economic index that reflects changes in the retail price of urban and rural goods. The adjustment and change of retail prices directly affects the living expenses of urban and rural residents and the state's fiscal revenue, affects residents' purchasing power and market supply-demand balance, and affects the ratio of consumption to accumulation. Therefore, calculating the retail price index can observe and analyze the above economic activities from one aspect.

Compiled by the National Bureau of Statistics

The adjustment and change of retail prices directly affects the living expenses of urban and rural residents and the state's fiscal revenue, affects residents' purchasing power and market supply-demand balance, and affects the ratio of consumption to accumulation. So calculate the retail price

The retail price is mainly the price of residents' purchase of consumer goods. According to the income level and consumption composition of urban and rural residents, it is compiled into urban retail prices.

Main content

include

If the current prediction object is in the state Ei (i = 1,2,3,4,5), then Pij describes that the current state Ei will turn to the state Ej (j = 1,2,3,4,5) in the future. possibility. According to the principle of maximum probability, the state corresponding to the largest of (Pi1, Pi2, Pi3, Pi4, Pi5) selected here is the prediction result.

Because the state of the retail price index of commodities in 2003 is slowly rising, the probability of reaching five states through one transfer is: P41 = 0.2000, P42 = 0.0000, P43 = 0.0000, P44 = 0.2000, P45 = 0.6000, by max { P41, P42, P43, P44, P45} = P45 = 0.6000

It can be seen that the retail price index of goods in 2004 will rise rapidly, that is, the retail price index of goods in 2004 will rise compared with 2003, and the increase index will increase by more than 5.

Retail price index

Comparing the predicted results with the actual results, from the original data in Figure 1, it can be concluded that the increase in the retail price index of goods in 2004 from 2003 was 357.5-350.5 = 7> 5, which is in line with the aforementioned rapid rise. It shows that the prediction result is accurate.

Figure 1 Retail price index

At the same time, in the Markov process, the state probabilities of different periods are represented by state vectors. And there is a formula (n) = (n-1) P, where P is a state transition matrix. According to this formula, the state vector of the retail price index of goods in 2004 can also be predicted: that is, the probability of a rapid rise in 2004 is 0.6000, which is greater than that in other states. Therefore, the possibility of a rapid rise in 2004 is greater, which is also consistent with the results of the previous forecast. Similarly, according to this formula, the state vector of the retail price index of commodities in recent years such as 2005 can also be predicted.

The results show that in 2005, the trend of the retail price index of commodities was 0.1945, the probability of a slow decline was 0.0500, the probability of a slow rise was 0.3982, and the probability of a rapid rise was 0.5564. Therefore, in 2005 It is more likely that the retail price index will continue to rise rapidly. After inspection, the prediction results for 2004 are consistent with the actual results, which shows that this method is reliable and accurate in predicting the retail price index of commodities, and the principle of the method is simple. It is a method following time series analysis and causality analysis. The latter is a scientific forecasting method. Any sequential dynamic system with no aftereffect can be predicted by this method. It has strong operability.

During the process of using this model to predict and analyze, the results obtained are interval predictions, but more states cannot be set to improve accuracy. This way, although high accuracy results cannot be obtained, the accuracy of the prediction is improved.

The above conclusions are derived under the assumption that the state transition probability is stable, that is, it is assumed that the initial state vector and the state transition probability matrix remain unchanged. If this condition is ignored, the retail price index values of the commodities in the following years will be unrestrictedly predicted in the example, which will cause a large error. Therefore, to ensure the accuracy of prediction, it is necessary to adjust the initial state vector and the state transition probability matrix from time to time according to the actual situation. Therefore, the Markov chain method is more suitable for near-term prediction. ^[2]