1. bookAHEAD OF PRINT
Détails du magazine
License
Format
Magazine
eISSN
2444-8656
Première parution
01 Jan 2016
Périodicité
2 fois par an
Langues
Anglais
Accès libre

Enterprise financial strategy and performance management analysis based on principal component analysis

Publié en ligne: 15 Jul 2022
Volume & Edition: AHEAD OF PRINT
Pages: -
Reçu: 26 Feb 2022
Accepté: 25 Apr 2022
Détails du magazine
License
Format
Magazine
eISSN
2444-8656
Première parution
01 Jan 2016
Périodicité
2 fois par an
Langues
Anglais
Principle analysis of principal component analysis

To put it simply, principal component analysis is a mathematical transformation method that transforms a given set of related variables into another set of unrelated variables in linear transformation, and rearranges the new variables in descending order of decreasing variance. Under the condition that the overall variance remains unchanged, the first variable has the maximum variance, which is regarded as the first component. The variance value ranked second, and the variable not associated with the first variable was regarded as the second principal component. By analogy, a variable has a principal component. There are multiple variables in the data, and there is correlation between the variables, then there may be problems such as variable duplication or information intersection, and there are many subsequent problems caused by the unprocessed data[1].

Definition 1

Assuming that there are n sample units and all sample units have P indicators, construct the matrix X= (Xij) n*p, where represents the JTH financial indicator of the ith company, as shown below[2].: X=[X11X1pXn1Xnp] X = \left[{\matrix{{{X_{11}} \ldots {X_{1p}}} \hfill \cr \ldots \hfill \cr {{X_{n1}} \ldots {X_{np}}} \hfill \cr}} \right]

In order to achieve standardized data, it is necessary to solve the differences between the original financial data and build a standardized matrix, as shown below: Z(Zij)n*p Z{\left({{Z_{ij}}} \right)_{n*p}}

Theorem 1

Standardized data information is regarded as the basis, and statistical indicators of the close relationship between standardized data are presented after the covariance matrix is constructed. The larger the value is, the more principal component analysis is required. The matrix formula of z-correlation coefficient is as follows: Rij=1nk=1nZjkZkj(i,j=1,2p) {R_{ij}} = {1 \over n}\sum\nolimits_{k = 1}^n {ZjkZkj\left({i,\,j = 1,2 \ldots p} \right)}

The equation for calculating the eigenvalue and eigenvector of R is as follows: |RλI|=0 \left| {R - \lambda I} \right| = 0

Obtaining characteristic values λj(j = 1, 2, … p) The specific ordering of, can define the formula of feature vector U: Rμi=λiμi R\mu i = \lambda i\mu i

Proposition 2

Generally speaking, the first principal component contains the largest amount of information, and F1 represents the first component. Only when the information contained in the first principal component cannot fully explain the original variable, the second component can be selected, and there is no correlation between the two. The principal component score is calculated as follows: F1=μ11x1+μ12x2++μ1pxpF2=μ21x1+μ22x2++μ2pxpFm=μm1x1+μm2x2++μmpxp \matrix{{{F_1} = {\mu _{11}}{x_1} + {\mu _{12}}{x_2} + \ldots + {\mu _{1p}}{x_p}} \hfill \cr {{F_2} = {\mu _{21}}{x_1} + {\mu _{22}}{x_2} + \ldots + {\mu _{2p}}{x_p}} \hfill \cr \ldots \hfill \cr {{F_m} = {\mu _{m1}}{x_1} + {\mu _{m2}}{x_2} + \ldots + {\mu _{mp}}{x_p}} \hfill \cr}

There is a direct proportional relationship between the variance contribution rate of the principal component and the information contained. The larger the actual value is, the more information contained is. The variance contribution rate of the ith principal component is as follows: αi=λii=1pλiG(k)=i=1kλii=1pλi \matrix{{{\alpha _i} = {{{\lambda _i}} \over {\sum\nolimits_{i = 1}^p {{\lambda _i}}}}} \hfill \cr {G\left(k \right) = {{\sum\nolimits_{i = 1}^k {{\lambda _i}}} \over {\sum\nolimits_{i = 1}^p {{\lambda _i}}}}} \hfill \cr}

Lemma 3

By multiplying the principal component score by the variance contribution rate, the weight value can be obtained and the composite score for each sample can be obtained.

In the early 1940s, American statistician stone using the principal component analysis (pca), to the United States during the period of 1929 to 1938, the data information of system research, the variable elements of original research involves 17, and principal component analysis (pca) only need three variables, including the total income, the rate of change of total revenue, economic development and the trend of decline, The actual calculation accuracy can reach 97.4%. Based on the research results of this event, it can be seen that principal component analysis is to comprehensively simplify the actual multi-variable cross-section data table under the principle of minimizing the loss of data information. In other words, it is to reduce the dimension of high-dimensional variable space.[4]

Suppose we are discussed the practical problems contained in the p index, the index as a p a random variables, principal component analysis is to the index, converted to discuss the linear combination of multiple indexes, and these new indicators, need according to the principle of reserves the main information, fully show the information of original indexes, and independent of each other.

Conjecture 5 Principal component analysis usually calculates the linear combination Fi of the original index, and the specific formula is as follows: F1=μ11X1+μ21X2++μp1XpF2=μ12X1+μ22X2++μp2XpFp=μ1pX1+μ2pX2++μppXp \matrix{{{F_1} = {\mu _{11}}{X_1} + {\mu _{21}}{X_2} + \ldots + {\mu _{p1}}{X_p}} \hfill \cr {{F_2} = {\mu _{12}}{X_1} + {\mu _{22}}{X_2} + \ldots + {\mu _{p2}}{X_p}} \hfill \cr \ldots \hfill \cr {{F_p} = {\mu _{1p}}{X_1} + {\mu _{2p}}{X_2} + \ldots + {\mu _{pp}}{X_p}} \hfill \cr}

And this study meets the following conditions:

First, the sum of the square coefficients of all principal components is 1, and the specific formula is: u1i2+u2i2++upi2=1 u_{1i}^2 + u_{2i}^2 + \ldots + u_{pi}^2 = 1

Secondly, the principal components are independent of each other and there is no overlapping information. The specific formula is as follows: Cov(Fi,Fj)=0,ij,i,j=1,2,,p Cov\left({{F_i},{F_j}} \right) = 0,i \ne j,i,j = 1,2, \ldots,p

Finally, the variance of principal components will decrease successively, and their importance will also decrease. The corresponding formula is as follows: Var(F1)Var(F2)Var(Fp) Var\left({{F_1}} \right) \ge Var\left({{F_2}} \right) \ge \ldots \ge Var\left({{F_p}} \right)

According to the geometric interpretation diagram of the principal component analysis shown in FIG. 1 below, if the x1 and x2 axes are first shifted and θ angles are rotated counterclockwise, new coordinate axes F1 and F2 can be obtained, which are then two new variables.

FIG. 1

Geometric interpretation of principal component analysis

Example 6. According to the formula analysis of rotation transformation, the details are as follows: {y1=x1cosθ+x2sinθy1=x1sinθ+x2cosθ \left\{{\matrix{{{y_1} = {x_1}\cos \theta + {x_2}\sin \theta} \hfill \cr {{y_1} = - {x_1}\sin \theta + {x_2}\cos \theta} \hfill \cr}} \right. (y1y2)=(cosθsinθsinθcosθ)(x1x2)=Ux \left({\matrix{{{y_1}} \hfill \cr {{y_2}} \hfill \cr}} \right) = \left({\matrix{{\cos \,\theta \,\sin \,\theta} \hfill \cr {- \sin \,\theta \,\cos \,\theta} \hfill \cr}} \right)\left({\matrix{{{x_1}} \hfill \cr {{x_2}} \hfill \cr}} \right) = {U^{'}}x

In the above formula, U ‘represents the rotation transformation matrix, which belongs to the orthogonal matrix, and it can be obtained: U=U1,UU=1 {U^{'}} = {U^{- 1}},\,{U^{'}}U = 1

The purpose of rotation transformation is to maximize the discretization degree of N sample points along the F1 axis, in other words, to maximize the variance of variable F1. The variable F1 represents the vast majority of the original data. When studying corporate financial strategy and performance management, there will be no significant change even if the variable F2 is not studied. The rotation changes allow more information to be concentrated in axis F1, thus reducing the information contained in the data.[5]

Example 6. By studying the derivation properties of the principal components, it can be known that assuming A represents the real symmetric matrix of order P, the orthogonal matrix U can be defined, and thus: U1AU=[λ1000λ2000λp]p×p {U^{- 1}}AU = {\left[{\matrix{{{\lambda _1}0 \ldots 0} \hfill \cr {0{\lambda _2} \ldots 0} \hfill \cr \ldots \hfill \cr {00 \ldots {\lambda _p}} \hfill \cr}} \right]_{p \times p}}

In the above formula, λi, i = 1, 2, …, p Represents the characteristic root of A.

The unit eigenvector corresponding to the eigenroots of the above matrix is U1... Up, Assuming that U=(u1,up)=[u11u12u1pu21u22u2pup1up2upp] U = \left({{u_1}, \ldots {u_p}} \right) = \left[{\matrix{{{u_{11}}{u_{12}} \ldots {u_{1p}}} \hfill \cr {{u_{21}}{u_{22}} \ldots {u_{2p}}} \hfill \cr \ldots \hfill \cr {{u_{p1}}{u_{p2}} \ldots {u_{pp}}} \hfill \cr}} \right]

It can be clear that the real symmetric matrix A represents the orthogonal relationship between the corresponding feature vectors of different features, and it can be obtained as follows: UU=UU=I {U^{'}}U = U{U^{'}} = I

Open Problem 8. From the first principal component, assume that the covariance matrix of X is: x=[σ12σ12σ1pσ21σ22σ2pσp1σp2σp2] \sum\nolimits_x {= \left[{\matrix{{\sigma _1^2{\sigma _{12}} \ldots {\sigma _{1p}}} \hfill \cr {{\sigma _{21}}\sigma _2^2 \ldots {\sigma _{2p}}} \hfill \cr \ldots \hfill \cr {{\sigma _{p1}}{\sigma _{p2}} \ldots \sigma _p^2} \hfill \cr}} \right]}

Because it represents a nonnegative definite symmetric matrix, it can be obtained by using the knowledge of linear algebra, and there must be an orthogonal matrix U, thus: UXU=[λ100λp] {U^{'}}\sum\nolimits_X {U = \left[{\matrix{{{\lambda _1}0} \hfill \cr \ldots \hfill \cr {0{\lambda _p}} \hfill \cr}} \right]}

Example 7. In the above formula, λ1, λ2,…, λp On behalf of the ∑x Eigenroots, you can assume that λ1λ2 ≥ … ≥ λp. And U is an orthogonal matrix formed according to the eigenvectors corresponding to the characteristic roots, as shown below: U=(u1,,up)=[u11u12u1pu21u22u2pup1up2upp]Ui=(u1i,u2i,,upi)i=1,2,,p \matrix{{U = \left({{u_1}, \ldots,{u_p}} \right) = \left[{\matrix{{{u_{11}}{u_{12}} \ldots {u_{1p}}} \hfill \cr {{u_{21}}{u_{22}} \ldots {u_{2p}}} \hfill \cr \ldots \hfill \cr {{u_{p1}}{u_{p2}} \ldots {u_{pp}}} \hfill \cr}} \right]} \hfill \cr {{U_i} = \left({{u_{1i}},{u_{2i}}, \ldots,{u_{pi}}} \right)'\,i = 1,2, \ldots,p} \hfill \cr}

Based on the imperfect information of the first principal component, we should search for the second principal component. Under constraint cov(F1, F2) = 0 Below, find the second component as shown below: F2=u12X1++up2Xp {F_2} = {u_{12}}{X_1} + \ldots + {u_{p2}}{X_p}

Because in accordance with cov(F1,F2)=cov(u1x,u2x)=u2u1=λ1u2u1=0 {\mathop{\rm cov}} \left({{F_1},{F_2}} \right) = {\mathop{\rm cov}} \left({u_1^{'}x,u_2^{'}x} \right) = u_2^{'}\sum {{u_1} = {\lambda _1}u_2^{'}{u_1} = 0}

This condition, so can be obtained: u2u1=0 u_2^{'}{u_1} = 0

Then for the p-dimension vector U2, it can be obtained: V(F2)=u2u2=i=1pλiu2uiuiu2=i=1p(u2ui)2λ2i=2p(u2ui)2=λ2i=1pu2uiuiu2=λ2u2UUu2=λ2u2u2=λ2 \matrix{{V\left({{F_2}} \right) = u_2^{'}\sum {{u_2} = \sum\limits_{i = 1}^p {{\lambda _i}u_2^{'}{u_i}u_i^{'}{u_2} = \sum\limits_{i = 1}^p {{{\left({u_2^{'}{u_i}} \right)}^2} \le {\lambda _2}\sum\limits_{i = 2}^p {{{\left({u_2^{'}{u_i}} \right)}^2}}}}}} \hfill \cr {= {\lambda _2}\sum\limits_{i = 1}^p {u_2^{'}{u_i}u_i^{'}{u_2}}} \hfill \cr {= {\lambda _2}u_2^{'}UU'{u_2} = {\lambda _2}u_2^{'}{u_2} = {\lambda _2}} \hfill \cr}

So assume a curvilinear change: F2=u12X1+u22X2++up2Xp {F_2} = {u_{12}}{X_1} + {u_{22}}{X_2} + \ldots + {u_{p2}}{X_p}

Then F2 will have the second largest variance, and so on: F1=μ11X1+μ21X2++μp1XpF2=μ12X1+μ22X2++μp2XpFp=μ1pX1+μ2pX2++μppXp \matrix{{{F_1} = {\mu _{11}}{X_1} + {\mu _{21}}{X_2} + \ldots + {\mu _{p1}}{X_p}} \hfill \cr {{F_2} = {\mu _{12}}{X_1} + {\mu _{22}}{X_2} + \ldots + {\mu _{p2}}{X_p}} \hfill \cr \ldots \hfill \cr {{F_p} = {\mu _{1p}}{X_1} + {\mu _{2p}}{X_2} + \ldots + {\mu _{pp}}{X_p}} \hfill \cr}

The corresponding matrix form is: F=UXU=(u1,,up)=[u11u12u1pu21u22u2pup1up2upp]X=(X1,X2,,Xp) \matrix{{F = {U^{'}}X} \hfill \cr {U = \left({{u_1}, \ldots,{u_p}} \right) = \left[{\matrix{{{u_{11}}{u_{12}} \ldots {u_{1p}}} \hfill \cr {{u_{21}}{u_{22}} \ldots {u_{2p}}} \hfill \cr \ldots \hfill \cr {{u_{p1}}{u_{p2}} \ldots {u_{pp}}} \hfill \cr}} \right]} \hfill \cr {X = \left({{X_1},{X_2}, \ldots,{X_p}} \right)'} \hfill \cr}

Financial performance evaluation system based on an enterprise
Selection of evaluation indicators

Since the selection of evaluation indicators directly affects the overall evaluation results, the influence of evaluation indicators on the results should be fully studied in the selection. In this paper, when an enterprise chooses its financial indicators, it will study the four abilities of the enterprise, the first is the solvency, the second is the profitability, the third is the operation ability, and the last is the growth ability. The actual selected financial indicators are shown in Table 1 below:[6].

The meaning of financial indicators

Indicators Meaning
Return on equity X1 Net profit/average balance of shareholders’ equity *100%
Net margin on sales X2 Net profit/net operating income *100%
Gross margin on sales X3 Gross profit on sales/net operating income *100%
Asset-liability ratio X4 Liabilities/Assets *100%
Current ratio X5 Current assets/current liabilities
Quick ratio X6 (Current assets - Inventory)/current liabilities
Inventory turnover X7 Cost of sales/average inventory balance
Accounts receivable turnover X8 Net credit sales/average balance of accounts receivable
Profit growth rate X9 (Net profit of current period - net profit of the same period of last year)/Net profit of the same period of last year *100%
Growth rate of assets X10 (Asset ending value of current period - initial value of asset of current period)/Initial value of asset of current period *100%
Evaluation System

First, the possibility of principal component analysis of the original data should be judged. Both bartlett test and K MO test can be used to judge whether the original data conforms to the principal component analysis method, and its application principle is to check the correlation between variables. The larger the value of K mo obtained, the greater the correlation between variables is proved, so it is very suitable for principal component analysis. Generally speaking, there is a definite limit to this value. When the value is above 0.5, principal component analysis can be used, otherwise it is not suitable. According to the analysis of the test results shown in the table below, the value of the selected variables obtained in this study is equal to 0.630, which exceeds the defined value of the applicable method. Therefore, principal component analysis can be used in this study. However, the significance of Bartlett's sphericity test is equal to zero, which proves that the value is lower than the actual limit value of 0.05, proving that a small number of principal components can be obtained from the proposed variable.

Test results

0KMO sampling suitability quantity .630
Badlett sphericity test The approximate chi-square 355.037
Degrees of freedom 46
significant .000

Second, make clear the degree of commonality of variables. The initial value and proposed value of this content can be listed according to variance in the software calculation, which can intuitively present the information number contained in the original variable and the information number after extraction, and intuitively determine whether the proposed principal component can explain the large amount of information of the original data. According to the analysis of the common factor variance table as shown in Table 3 below, except for the low degree of commonality of the variables of asset growth rate, the commonness of the variables of profit growth rate is even 97.4%, which proves that the obtained principal components can directly explain the original variables.

Common factor variance table

The initial Extract
X1 1.000 .846
X2 1.000 .811
X3 1.000 .877
X4 1.000 .853
X5 1.000 .911
X6 1.000 .804
X7 1.000 .867
X8 1.000 .892
X9 1.000 .973
X10 1.000 .562

Third, rubble map. The horizontal coordinate contains 10, corresponding to each serial number of the principal component studied in this paper. The vertical axis refers to the size of the characteristic value, and the value closer to the horizontal axis, the smaller it is. This graph can intuitively show the characteristics of which principal component, and the largest value represents the stronger the actual importance. Based on the results shown in Figure 2 below, the first four main variables can be considered important in explaining all the variables.

FIG. 2

Lithotripsy

Fourth, extract the principal component. In the reduction analysis, the conclusion of the total variance interpretation table can show whether the variance contribution rate of the principal component and the cumulative variance contribution rate meet the requirements. Generally speaking, if the cumulative variance contribution rate exceeds 70%, it means that it can prove the core information of the original index. According to the research results shown in the following table, the cumulative variance contribution rate of the first four principal components reached 80.710%.

Total variance interpretation table

Initial eigenvalue variance percent Extract load square and variance percentage Rotational load square and variance percent
composition A total of than Cumulative % A total of than Cumulative % A total of than Cumulative %
1 3.861 36.999 36.999 3.862 36.999 36.999 2.720 24.939 24.939
2 2.112 19.277 56.376 2.111 19.377 56.376 2.403 18.967 51.706
3 1.278 12.769 69.145 1.279 12.769 69.145 1.933 17.069 68.775
4 1.047 11.565 80.710 1.047 11.565 80.710 1.096 11.935 80.710
5 .694 6.998 88.391
6 .383 3.847 93.307
7 .285 2.698 95.979
8 .154 1.530 97.533
9 .128 1.268 98.684
10 .113 1.154 100.000

Fifth, independence test. According to the analysis of component score covariance matrix table as shown in Table 5 below, the difference of covariance between the four components is 0, which proves that there is no correlation between them.

Matrix table of component score covariance

Composition 1 2 3 4
1 1.000 .000 .000 .000
2 .000 1.000 .000 .000
3 .000 .000 1.000 .000
4 .000 .000 .000 1.000

Sixth, principal component function analysis. The principal component of solvency is regarded as F1, the principal component of profitability as F2, the principal component of operating capacity as F3, and the principal component of growth capacity as F4. According to the score coefficient matrix of standardized processing of original variables, the matrix of calculation function is shown as Table 6 follows:

Coefficient matrix table of component scores

1 2 3 4
X1 .079 −.302 −.321 .167
X2 −.017 .363 .004 −.114
X3 .225 .255 .208 .089
X4 −.339 −.023 −.003 .007
X5 .401 .411 .191 .064
X6 .396 .115 .253 .030
X7 .113 −.008 −.493 .057
X8 −.004 −.289 .377 −.177
X9 .104 .216 .098 .899
X10 −.043 −.109 .597 .199

The corresponding score calculation function is: F=0.098*ReturnofEquity0.017*Netprofitmarginonsales+0.225*Grossprofitmarginonsales0.339*DebttoAssetsratio+0.401*Currentratio+0.396*DebttoAssetsratio+0.401*Currentratio+0.396*Quickratio+0.113*Inventoryturnover0.004*Accountsreceivableturnover+0.104*Profitgrowthrate0.043*Assetgrowthrate \matrix{{F = 0.098*\,{\rm{Return}}\,{\rm{of}}\,{\rm{Equity}} - 0.017*{\rm{Net}}\,{\rm{profit}}\,{\rm{margin}}\,{\rm{on}}\,{\rm{sales}}\, + \,0.225*{\rm{Gross}}\,{\rm{profit}}\,{\rm{margin}}\,{\rm{on}}} \hfill \cr {{\rm{sales}} - 0.339*{\rm{Debt}} - {\rm{to}} - {\rm{Assets}}\,{\rm ratio}\, + \,0.401*\,{\rm{Current}}\,{\rm{ratio}}\, + \,0.396*{\rm{Debt}} - {\rm{to}} - {\rm{Assets}}\,{\rm{ratio}}\, +} \hfill \cr {0.401*\,{\rm{Current}}\,{\rm{ratio}}\, + \,0.396*\,{\rm{Quick}}\,{\rm{ratio}}\, + \,0.113\,*\,{\rm{Inventory}}\,{\rm{turnover}} - 0.004*\,{\rm{Accounts}}\,{\rm{receivable}}} \hfill \cr {{\rm{turnover}} + 0.104*\,{\rm Profit\,growth\,rate}\, - \,0.043*\,{\rm Asset\,growth\,rate}} \hfill \cr}

Finally, after summary and analysis, the variance contribution rate of the four principal components is regarded as the coefficient of the evaluation score, and the weight is linearly weighted to the comprehensive, thus obtaining the following function: F=(24.939%*F1+18.967%F2+17.069%*F3+11.935%*F4)/80.71%=0.309*F1+0.235*F2+0.2115*F3+0.1479*F4 \matrix{{F = \left({24.939\% *F1 + 18.967\% F2 + 17.069\% *F3 + 11.935\% *F4} \right)/80.71\%} \hfill \cr {= 0.309*F1 + 0.235*F2 + 0.2115*F3 + 0.1479*F4} \hfill \cr}\matrix{{F = \left({24.939\% *F1 + 18.967\% F2 + 17.069\% *F3 + 11.935\% *F4} \right)/80.71\%} \hfill \cr {= 0.309*F1 + 0.235*F2 + 0.2115*F3 + 0.1479*F4} \hfill \cr}

Financial performance evaluation and strategic management analysis based on an enterprise
Principal component analysis results

According to the ranking results of the research enterprise among the peer listed enterprises, the details are shown in Table 7 below. An enterprise ranked 11th in the comprehensive score last year, which proves that the financial performance of the enterprise is at a low level among these companies. In addition to the principal component ranking of the growth ability, the principal components of the other three capabilities are in the middle and lower reaches of the industry. For example, in the principal component analysis of profitability, the average score of the enterprise studied in this paper is −0.0633, far lower than the average score of other enterprises, and accounts for 18.967% in the four principal component analysis, thus proving that the principal component analysis of profitability is also the main factor affecting the financial performance of enterprises. In the principal component analysis of enterprise growth ability, the actual score can reach 0.145, which is in the middle and lower reaches of the industry ranking, but there is a large room for improvement of the actual growth ability[7.8].

Score statistics of sample enterprises in 2020

The company referred to as F1 F2 F3 F4 F Comprehensive ranking
Sanhua intellectual control 4.9681 2.2115 4.5898 0.026 3.0294 1
Han clock precision machine 4.0291 3.0532 3.3143 1.0355 2.8166 2
Good try to 3.5571 3.4277 2.561 0.0784 2.3579 3
Double liang energy-saving 4.5730 2.6331 3.7981 −2.9552 2.3981 4
Ice cold or hot 2.8285 1.7479 5.1566 −0.7148 2.2697 5
The vic 2.6417 3.2867 3.0437 0.0047 2.2331 6
Moon co 3.0747 2.8874 3.0775 −2.092 1.9701 7
In accordance with the meters 0.7784 0.1396 4.6653 0.0643 1.2696 8
Company 0.9165 0.2155 1.7859 0.3764 0.7672 9
The peak of electronic 2.9744 −0.0583 2.7993 −6.3473 0.5587 10
The snowman shares 0.7498 −0.0268 0.9747 −0.8154 0.3109 11
Strategic management countermeasures for enterprise performance

First of all, to build a sound financial performance evaluation department. During financial performance appraisals, most of the data information is derived from financial statements, so it is considered the business of the accounting department. However, from the perspective of the development of enterprise organizational structure, based on the requirements of enterprise strategic management, the establishment of a comprehensive financial center, including the management of securities department, audit department, financial department and other staff, can effectively integrate and use department resources, and jointly participate in the evaluation of enterprise financial performance. The specific structure is shown in Figure 3 below:

Figure 3

Organizational chart of the enterprise

Secondly, it is necessary to strengthen the professional quality of enterprise performance appraisal personnel. Financial performance evaluation work to by finance department personnel to carry out, so if the finance department personnel lack of professional knowledge and professional quality, so the actual financial performance evaluation work is bound to be problems, such as unscientific indicators, financial performance evaluation result is not accurate, etc., finally it is difficult to achieve the requirements of performance management. Therefore, enterprises should conduct comprehensive assessment when recruiting professionals, pay attention to regular training of internal financial personnel, encourage and support employees to learn professional knowledge and skills in salty fish time.

Finally, to strengthen the management level of enterprise financial performance. On the one hand, we must constantly optimize the capital structure. The capital structure of an enterprise will be affected by internal and external factors, among which internal factors involve enterprise scale, profit level, site capacity, asset structure and other contents. External factors include national economy, economic cycle, industry competition and so on. Generally speaking, the capital structure should put forward two levels of standards, one should be set according to their own development, and the other should pay attention to the target of leading enterprises in the industry, pay attention to the concept of building dynamic capital structure, so as to increase the current assets of enterprises, control the proportion of current liabilities, and control the debt ratio within a reasonable range. On the other hand, we should put forward a unified asset management system. It is necessary to not only put forward a perfect post responsibility system for fixed assets, but also arrange special personnel for control during use, integrate relevant maintenance indicators into performance assessment, and conduct regular or irregular inspection and management. Meanwhile, regular maintenance should be carried out for fixed equipment, and application equipment should be updated and optimized in time, so as to improve the application efficiency and overall operation level of fixed assets.[9]

Conclusion

To sum up, according to the research results and performance evaluation of enterprise financial system, find the problems existing in the enterprise financial performance management, optimizing the performance evaluation of enterprise financial system, promoting the comprehensive qualities of the performance evaluation of financial personnel, strengthen the application research of principal component analysis (pca), help enterprise during the period of construction management into a debt crisis, and to obtain more economic benefits in the practice exploration.

FIG. 1

Geometric interpretation of principal component analysis
Geometric interpretation of principal component analysis

FIG. 2

Lithotripsy
Lithotripsy

Figure 3

Organizational chart of the enterprise
Organizational chart of the enterprise

Score statistics of sample enterprises in 2020

The company referred to as F1 F2 F3 F4 F Comprehensive ranking
Sanhua intellectual control 4.9681 2.2115 4.5898 0.026 3.0294 1
Han clock precision machine 4.0291 3.0532 3.3143 1.0355 2.8166 2
Good try to 3.5571 3.4277 2.561 0.0784 2.3579 3
Double liang energy-saving 4.5730 2.6331 3.7981 −2.9552 2.3981 4
Ice cold or hot 2.8285 1.7479 5.1566 −0.7148 2.2697 5
The vic 2.6417 3.2867 3.0437 0.0047 2.2331 6
Moon co 3.0747 2.8874 3.0775 −2.092 1.9701 7
In accordance with the meters 0.7784 0.1396 4.6653 0.0643 1.2696 8
Company 0.9165 0.2155 1.7859 0.3764 0.7672 9
The peak of electronic 2.9744 −0.0583 2.7993 −6.3473 0.5587 10
The snowman shares 0.7498 −0.0268 0.9747 −0.8154 0.3109 11

Total variance interpretation table

Initial eigenvalue variance percent Extract load square and variance percentage Rotational load square and variance percent
composition A total of than Cumulative % A total of than Cumulative % A total of than Cumulative %
1 3.861 36.999 36.999 3.862 36.999 36.999 2.720 24.939 24.939
2 2.112 19.277 56.376 2.111 19.377 56.376 2.403 18.967 51.706
3 1.278 12.769 69.145 1.279 12.769 69.145 1.933 17.069 68.775
4 1.047 11.565 80.710 1.047 11.565 80.710 1.096 11.935 80.710
5 .694 6.998 88.391
6 .383 3.847 93.307
7 .285 2.698 95.979
8 .154 1.530 97.533
9 .128 1.268 98.684
10 .113 1.154 100.000

Common factor variance table

The initial Extract
X1 1.000 .846
X2 1.000 .811
X3 1.000 .877
X4 1.000 .853
X5 1.000 .911
X6 1.000 .804
X7 1.000 .867
X8 1.000 .892
X9 1.000 .973
X10 1.000 .562

Coefficient matrix table of component scores

1 2 3 4
X1 .079 −.302 −.321 .167
X2 −.017 .363 .004 −.114
X3 .225 .255 .208 .089
X4 −.339 −.023 −.003 .007
X5 .401 .411 .191 .064
X6 .396 .115 .253 .030
X7 .113 −.008 −.493 .057
X8 −.004 −.289 .377 −.177
X9 .104 .216 .098 .899
X10 −.043 −.109 .597 .199

Matrix table of component score covariance

Composition 1 2 3 4
1 1.000 .000 .000 .000
2 .000 1.000 .000 .000
3 .000 .000 1.000 .000
4 .000 .000 .000 1.000

The meaning of financial indicators

Indicators Meaning
Return on equity X1 Net profit/average balance of shareholders’ equity *100%
Net margin on sales X2 Net profit/net operating income *100%
Gross margin on sales X3 Gross profit on sales/net operating income *100%
Asset-liability ratio X4 Liabilities/Assets *100%
Current ratio X5 Current assets/current liabilities
Quick ratio X6 (Current assets - Inventory)/current liabilities
Inventory turnover X7 Cost of sales/average inventory balance
Accounts receivable turnover X8 Net credit sales/average balance of accounts receivable
Profit growth rate X9 (Net profit of current period - net profit of the same period of last year)/Net profit of the same period of last year *100%
Growth rate of assets X10 (Asset ending value of current period - initial value of asset of current period)/Initial value of asset of current period *100%

Test results

0KMO sampling suitability quantity .630
Badlett sphericity test The approximate chi-square 355.037
Degrees of freedom 46
significant .000

Ying Qin. Analysis on current Situation and Countermeasures of Enterprise Financial Strategic Management -- Taking C Machine Tool Company as an example [J]. Enterprise reform and management, 2021 (22) : 189–190. The DOI: 10.13768/j.carolcarrollnkicn11-3793/f.,2021.2309. QinYing Analysis on current Situation and Countermeasures of Enterprise Financial Strategic Management -- Taking C Machine Tool Company as an example [J] Enterprise reform and management 2021 22 189 190 10.13768/j.carolcarrollnkicn11-3793/f.,2021.2309 Ouvrir le DOISearch in Google Scholar

Ping Zhang. Corporate Financial Strategy under asset-light Operation Mode -- TAKING XX Enterprise as an example [J]. Finance, 2021 (32) : 147–148. The DOI: 10.19887/j.carolcarrollnkicn11-4098/f2021.32.072. ZhangPing Corporate Financial Strategy under asset-light Operation Mode -- TAKING XX Enterprise as an example [J] Finance 2021 32 147 148 10.19887/j.carolcarrollnkicn11-4098/f2021.32.072 Ouvrir le DOISearch in Google Scholar

Qian Yu. On the research framework construction of financial strategic management in the growth stage of enterprises [J]. Modernization, 2021 (20) : 156–158. The DOI: 10.14013/j.carolcarrollnkiSCXDH.2021.20.057. YuQian On the research framework construction of financial strategic management in the growth stage of enterprises [J] Modernization 2021 20 156 158 10.14013/j.carolcarrollnkiSCXDH.2021.20.057 Ouvrir le DOISearch in Google Scholar

Ting Lin. Application analysis of rolling Budget in Aviation Maintenance Industry [J]. Administrative Assets and Finance, 2021(20): 73–74. LinTing Application analysis of rolling Budget in Aviation Maintenance Industry [J] Administrative Assets and Finance 2021 20 73 74 Search in Google Scholar

yan-ling Chen. Medical equipment enterprise financial strategy formulation and implementation of [J]. Journal of enterprise reform and management, 2021 (20) : 188–189. The DOI: 10.13768/j.carolcarrollnkicn11-3793/f2021.2098. Chenyan-ling Medical equipment enterprise financial strategy formulation and implementation of [J] Journal of enterprise reform and management 2021 20 188 189 10.13768/j.carolcarrollnkicn11-3793/f2021.2098 Ouvrir le DOISearch in Google Scholar

Lifen Yang, Xiang Zhu. Research on financial strategy of agricultural enterprises from the perspective of value chain -- taking a certain enterprise as an example [J]. Quality and Market, 2021(20): 28–30. YangLifen ZhuXiang Research on financial strategy of agricultural enterprises from the perspective of value chain -- taking a certain enterprise as an example [J] Quality and Market 2021 20 28 30 Search in Google Scholar

kai Miao. Financial management objectives and corporate Financial strategy choice [J]. Taxation, 201, 15(29): 79–80. Miaokai Financial management objectives and corporate Financial strategy choice [J] Taxation 201, 15 29 79 80 Search in Google Scholar

Yanchao Deng. Financial strategy of broadcasting enterprises under the background of supply-side structural reform [J]. Business Culture, 2021(29): 96–97.] DengYanchao Financial strategy of broadcasting enterprises under the background of supply-side structural reform [J] Business Culture 2021 29 96 97 Search in Google Scholar

Xiaonan Li. Enterprise financial strategic risk analysis and control strategy exploration [J]. Accounting for Township Enterprises in China, 2021(09): 61–62. LiXiaonan Enterprise financial strategic risk analysis and control strategy exploration [J] Accounting for Township Enterprises in China 2021 09 61 62 Search in Google Scholar

Articles recommandés par Trend MD

Planifiez votre conférence à distance avec Sciendo