*3.4. Modeling*

This paper used the simultaneous equations method to test the relationship between RD investment and financially sustainable performance and performs 3SLS to build Models (3)–(6).

$$RD\_{i,t} = a\_0 + a\_1 STP\_{i,t} + a\_2 STP\_{i,t-1} + a\_3 STP\_{i,t-2} + \theta\_l \sum\_{i=1}^{i=l} C\_{i,t} + \mu\_{i,t} \tag{3}$$

$$RD\_{i,t} = a\_0 + a\_1 LTD\_{i,t} + a\_2 LTD\_{i,t-1} + a\_3 LTD\_{i,t-2} + \theta\_l \sum\_{i=1}^{i=l} C\_{i,t} + \mu\_{i,t} \tag{4}$$

$$STP\_{i,t} = \beta\_0 + \beta\_1 RD\_{i,t} + \beta\_2 RD\_{i,t-1} + \beta\_3 RD\_{i,t-2} + \theta\_m \sum\_{i=1}^{i=m} C\_{i,t} + \varepsilon\_{i,t} \tag{5}$$

*Energies* **2020**, *13*, 2332

$$LTD\_{i,t} = \beta\_0 + \beta\_1 RD\_{i,t} + \beta\_2 RD\_{i,t-1} + \beta\_3 RD\_{i,t-2} + \theta\_m \sum\_{i=1}^{i=m} \mathcal{C}\_{i,t} + \varepsilon\_{i,t} \tag{6}$$

To examine the impact of executive incentive on the relationship between RD investment and financially sustainable performance and avoid the endogenous interaction between RD investment and financially sustainable performance, we used the dependent variable of the lag period and the interaction term between RD investment and executive incentive to explore the impact of executive incentive. Then, Models (7) and (8) were created as follows.

$$\begin{aligned} STP\_{i,t} &= \gamma\_0 + \gamma\_1 RD\_{i,t-1} + \gamma\_2 RD\_{i,t-2} + \gamma\_3 SI\_{i,t-1} + \gamma\_4 EI\_{i,t-1} + \gamma\_5 IN\_{i,t-1} \\ &\ast Sl\_{i,t-1} + \gamma\_6 RD\_{i,t-1} \ast EI\_{i,t-1} + \partial\_n \sum\_{i=1}^{i=n} \mathbb{C}\_{i,t} + q\_{i,t} \end{aligned} \tag{7}$$

$$\begin{aligned} LTD\_{i,t} &= \gamma\_0 + \gamma\_1 RD\_{i,t-1} + \gamma\_2 RD\_{i,t-2} + \gamma\_3 SI\_{i,t-1} + \gamma\_4 EI\_{i,t-1} + \gamma\_5 RD\_{i,t-1} \\ &\ast SI\_{i,t-1} + \gamma\_6 RD\_{i,t-1} \ast EI\_{i,t-1} + \theta\_n \sum\_{i=1}^{i=n} \mathbb{C}\_{i,t} + \varphi\_{i,t} \end{aligned} \tag{8}$$

#### **4. Results, Nalyses and Discussion**

#### *4.1. Descriptive and Correlation Analysis*

The descriptive analysis results are shown in Table 3. To eliminate the influence of extreme values, this paper uses the Winsorize method to deal with continuous variables according to 1% and 99%. The results in Table 3 demonstrate that the mean of RD investment is 0.031, indicating that the RD investment of enterprises is insufficient. The mean of salary incentives is 6.14 million RMB, but the difference between the minimum value and the maximum value is large, which shows that there is a significant difference in executive incentives. The mean of equity incentive is 0.105, but there are also extreme cases in which managers have no shares or executives hold more than 0.721 shares. Therefore, Chinese energy enterprises have implemented different equity incentives.


**Table 3.** Descriptive analysis.

The correlation analysis results are shown in Table 4. The Spearman correlation coefficient between RD investment and financially sustainable performance (STP, LTD) is significantly positive at the levels of 1% and 5%, respectively, while Pearson correlation coefficient is not significantly negative, which indicates that RD investment and financially sustainable performance may not be a one-way relationship, and their relationship needs further study. In terms of executive incentives, salary incentives and equity incentives are positively related to short-term profitability and long-term development capacity, which shows that executive incentives are conducive to the improvement of financially sustainable performance. In addition, there is no significant correlation between salary incentives and equity incentives, which helps to examine the role of the two in Models (7) and (8). The other correlation coefficients in Table 4 are small, indicating that there is no multicollinearity between variables.


**Table 4.** Spearman and Pearson correlation analysis.

Note: The upper and lower triangle are Spearman and Pearson correlation coefficient, respectively; \*\*\*,\*\*,\*Significantat1%,5%,and10%levels,respectively.
