ALSTR-NLRE model is backwardly constructed. In the framework of LSTR-NLRE model, using Chinese economic data during the first quarter of 2000 and the third quarter of 2011, this paper positively analyzes the nonlinearity and asymmetry of Chinese monetary policy rule with LSTR, and positively analyzes and simulates the effects on indeterminacy based on threshold value and transfer speed. The backward Modeling process shows that, in order to make optimal monetary policy rule in the NLRE system consistent with nonlinear rule with LSTR, nonlinear Philips curve should be an exponential function. The positive analyses shows that: Chinese monetary policy has the characteristics of nonlinearity and asymmetry, and the threshold value is positively related with inflation variation. During the first quarter of 2000 and the third quarter of 2011, the threshold of inflation expectation is 3% , while from the fourth quarter of 2011 to the third quarter of 2016, the threshold of inflation expectation is 2. 33% . A reasonable target of inflation expectation is very important to the central bank. Furthermore, threshold value has a greater influence on indeterminacy than transfer speed. In the high regime, nonlinear monetary policy rule with LSTR is more likely to stabilize output gap than inflation. In the low regime, nonlinear monetary policy rule with LSTR is more likely to stabilize inflation and is more likely to lead to output indeterminacy with long policy lag. If output gap is highly weighted, the central bank should increase inflation expectation above threshold value to put monetary policy into the high regime. If inflation is highly weighted, the central bank should decrease inflation expectation below the threshold value to put monetary policy into the low regime.