1. bookAHEAD OF PRINT
Dettagli della rivista
License
Formato
Rivista
eISSN
2444-8656
Prima pubblicazione
01 Jan 2016
Frequenza di pubblicazione
2 volte all'anno
Lingue
Inglese
Accesso libero

Financial Risk Prediction and Analysis Based on Nonlinear Differential Equations

Pubblicato online: 15 Jul 2022
Volume & Edizione: AHEAD OF PRINT
Pagine: -
Ricevuto: 21 Apr 2022
Accettato: 17 Jun 2022
Dettagli della rivista
License
Formato
Rivista
eISSN
2444-8656
Prima pubblicazione
01 Jan 2016
Frequenza di pubblicazione
2 volte all'anno
Lingue
Inglese
Introduction

Timely and practical discovery of financial risks is significant for enterprises to gain an advantage in market competition. The convenient and accurate evaluation of enterprise financial risk in academia has become a hot research issue [1]. Researchers build a multivariate financial risk evaluation model based on functional equations. However, the evaluation accuracy of this model is low. Non-statistical evaluation methods are the hotspot of current economic evaluation research. Standard methods include decision trees, random forests, neural networks, genetic algorithms, support vector machines, etc. At present, most financial risk evaluation models rely solely on economic indicators to achieve risk prediction [2]. The non-financial indicators in the process of enterprise operation also have a more significant impact on financial risks. However, enterprise historical data is messy and disordered. The existing financial risk assessment methods are based on statistical assumptions. These models are difficult to adapt to the current characteristics of corporate financial data.

For this reason, this paper constructs an enterprise financial risk evaluation model based on grey system theory. The model can comprehensively consider the financial and non-financial indicators of the enterprise. This model takes full advantage of the gray system's ability to deal with messy data. At the same time, the system uses the particle swarm optimization algorithm improved by the chaotic model to optimize the gray model weight coefficient [3]. Finally, this paper uses the data of a real estate company to verify the validity of the evaluation model.

The basic theory of the grey system

Grey's system theory stems from the cognitive need for uncertainty in analyzing things. Its basic principle is based on the laws that appear in realizing the transformation of things based on sequence operators [4]. We generate regular sequences from the original irregular data. Then we use the differential equation solving method to carry out the objective and scientific prediction and analysis of the development trend of things. The basic modeling process of GM(1,1) is as follows.

Suppose the number of observations of the system characteristic is: Xi(0)=[xi(0)(1),xi(0)(2),,xi(0)(m)] X_i^{\left( 0 \right)} = \left[ {x_i^{\left( 0 \right)}\left( 1 \right),x_i^{\left( 0 \right)}\left( 2 \right), \cdots ,x_i^{\left( 0 \right)}\left( m \right)} \right]

We accumulate the observation series. The resulting cumulative number column is: Xi(1)=[xi(0)(1),k=12xi(0)(k),,k=1nxi(0)(k)] X_i^{\left( 1 \right)} = \left[ {x_i^{\left( 0 \right)}\left( 1 \right),\sum\limits_{k = 1}^2 {x_i^{\left( 0 \right)}\left( k \right), \cdots ,\sum\limits_{k = 1}^n {x_i^{\left( 0 \right)}\left( k \right)} } } \right]

We convert the discrete-time instants in the cumulative sequence Xi(1) X_i^{\left( 1 \right)} into continuous-time variables. i.e. Xi(1)=Xi(1)(t) X_i^{\left( 1 \right)} = X_i^{\left( 1 \right)}\left( t \right) . The arrangement X2(1),X3(1),,Xm(1) X_2^{\left( 1 \right)},X_3^{\left( 1 \right)}, \cdots ,X_m^{\left( 1 \right)} will affect the rate of change of X1(1) X_1^{\left( 1 \right)} during the system change [5]. The grey model differential equation we constructed can be expressed as: dX1(1)dt+aX1(1)=b2X2(1)+b3X3(1)++bmXm(1) {{dX_1^{\left( 1 \right)}} \over {dt}} + aX_1^{\left( 1 \right)} = {b_2}X_2^{\left( 1 \right)} + {b_3}X_3^{\left( 1 \right)} + \cdots + {b_m}X_m^{\left( 1 \right)} Where a and bj(j = 2,3, ⋯ m) are the model parameters of the differential equation. The differential equation of Eq. (3) can be transformed into the form of a system of linear equations: Yn=Bα^ {Y_n} = B\hat \alpha

In α^=(a,b1,b2,,bm)T \hat \alpha = {\left( {a,{b_1},{b_2}, \cdots ,{b_m}} \right)^T} Yn=[x1(0)(2),x1(0)(3),,x1(0)(m)]T {Y_n} = {\left[ {x_1^{\left( 0 \right)}\left( 2 \right),x_1^{\left( 0 \right)}\left( 3 \right), \cdots ,x_1^{\left( 0 \right)}\left( m \right)} \right]^T} B=(12[x1(1)(1)+x1(1)(2)],x2(1)(2),,xm(1)(1)) B = \left( { - {1 \over 2}\left[ {x_1^{\left( 1 \right)}\left( 1 \right) + x_1^{\left( 1 \right)}\left( 2 \right)} \right],x_2^{\left( 1 \right)}\left( 2 \right), \cdots ,x_m^{\left( 1 \right)}\left( 1 \right)} \right)

The parameter vector can be represented as: α^=(BTB)1BTYm \hat \alpha = {\left( {{B^T}B} \right)^{ - 1}}{B^T}{Y_m}

At this time, the approximate solution of the differential equation can be expressed as:

A cumulative subtraction calculation can restore the original sequence: x^1(0)(k+1)=x^1(1)(k+1)x^1(1)(k) \hat x_1^{\left( 0 \right)}\left( {k + 1} \right) = \hat x_1^{\left( 1 \right)}\left( {k + 1} \right) - \hat x_1^{\left( 1 \right)}\left( k \right)

The final error sequence of the system prediction result x^1(0)(k) \hat x_1^{\left( 0 \right)}\left( k \right) can be expressed as: ε(0)(k)=|X1(0)(k)x^1(0)(k)| {\varepsilon ^{\left( 0 \right)}}\left( k \right) = \left| {X_1^{\left( 0 \right)}\left( k \right) - \hat x_1^{\left( 0 \right)}\left( k \right)} \right|

The multi-parameter financial risk assessment model
Indicator system

The choice of corporate financial indicators significantly impacts the evaluation performance of corporate financial risks. There is no unified indicator system for corporate financial risks [6]. This paper analyzes the existing corporate financial risk evaluation index system at home and abroad. Enterprise financial risk evaluation indicators are summarized, tested, and analyzed. At this time, we constructed an enterprise financial risk evaluation index system consisting of 13 economic indicators and 11 non-financial indicators.

Financial indicators

Financial indicators are the key to evaluating corporate financial risks. This paper proposes 13 economic indicators from four aspects: enterprise operation risk, investment risk, financing risk, and cash flow. The index structure is shown in Figure 1.

Figure 1

Financial metrics

Non-financial indicators

The non-financial indicators play an essential role in evaluating financial risks. This paper constructs 11 non-financial indicators from two aspects of internal enterprise factors and external factors. The index structure is shown in Figure 2.

Figure 2

Non-financial metrics

Model Construction

There are many indicators involved in enterprise financial risk early warning. The prediction effect of our direct application of the traditional gray model is not good [7]. Therefore, we construct a multi-parameter enterprise financial risk evaluation model. We write the system of differential equations in matrix form: dX(1)(t)dt=PX(1)(t)+Q {{d{X^{\left( 1 \right)}}\left( t \right)} \over {dt}} = P{X^{\left( 1 \right)}}\left( t \right) + Q Where P and Q are system identification parameters: P=[a11a12a1na21a22a2nan1an2ann] P = \left[ {\matrix{ {{a_{11}}} & {{a_{12}}} & \cdots & {{a_{1n}}} \cr {{a_{21}}} & {{a_{22}}} & \cdots & {{a_{2n}}} \cr \vdots & \vdots & {} & \vdots \cr {{a_{n1}}} & {{a_{n2}}} & \cdots & {{a_{nn}}} \cr } } \right] Q=[b1,b2,,bn]T Q = {\left[ {{b_1},{b_2}, \cdots ,{b_n}} \right]^T}

The forward and backward differential forms of the differential equation of Eq. (11) can be expressed as: X(1)(t+Δt)X(1)(t)=PX(1)(t)+Q {X^{\left( 1 \right)}}\left( {t + \Delta t} \right) - {X^{\left( 1 \right)}}\left( t \right) = P{X^{\left( 1 \right)}}\left( t \right) + Q X(1)(t+Δt)X(1)(t)=PX(1)(t+Δt)+Q {X^{\left( 1 \right)}}\left( {t + \Delta t} \right) - {X^{\left( 1 \right)}}\left( t \right) = P{X^{\left( 1 \right)}}\left( {t + \Delta t} \right) + Q

Δt is the system unit time interval.

An improved gray system model of particle swarm based on chaos
Particle Swarm Optimization

There are L particles in the particle swarm in the D dimensional parameter optimization space [8]. The particle swarm algorithm uses the flight of particles to search for the optimal value. Assume that the flight speed of the i particle in the particle swarm is: Vi=[vi1vi2viD] {V_i} = \left[ {{v_{i1}}\;\;\;{v_{i2}}\;\;\; \cdots \;\;\;{v_{iD}}} \right]

The spatial position of the corresponding particle can be expressed as: Fi=[fi1fi2fiD] {F_i} = \left[ {{f_{i1}}\;\;\;{f_{i2}}\;\;\; \cdots \;\;\;{f_{iD}}} \right]

The spatial position of the particle during flight constitutes the potential solution space of particle swarm optimization. The function of the maximum fitness value that the i particle passes through during the flight is called the extreme individual value of the article: Gi=[gi1gi2giD] {G_i} = \left[ {{g_{i1}}\;\;\;{g_{i2}}\;\;\; \cdots \;\;\;{g_{iD}}} \right]

The position of the maximum fitness value among all particles is called the population extremum: G¯i=[g¯i1g¯i2g¯iD] {\bar G_i} = \left[ {{{\bar g}_{i1}}\;\;\;{{\bar g}_{i2}}\;\;\; \cdots \;\;\;{{\bar g}_{iD}}} \right]

The particle swarm updates the particle's flight speed and spatial position through the individual extremum and the population extremum: {Vik+1=λVik+c1r1(GikFik)+c2r2(G¯ikFik)Fik+1=Fik+Vik+1 \left\{ {\matrix{ {V_i^{k + 1} = \lambda V_i^k + {c_1}{r_1}\left( {G_i^k - F_i^k} \right) + {c_2}{r_2}\left( {\bar G_i^k - F_i^k} \right)} \hfill \cr {F_i^{k + 1} = F_i^k + V_i^{k + 1}} \hfill \cr } } \right.

The role of inertia weight is to balance particle swarm optimization's local and global optimization ability [9].

Chaos Model

The chaotic state is sensitive to the initial finger and has the characteristics of universality and convenience [10]. The commonly used chaotic sequences include Logistic mapping, positive H mapping, Tent mapping, etc. Logistic mapping has stable performance in many chaotic systems and has extensive applications. z = (z1, z2, ⋯, zn1) vector A has dimension n1. Its expression is: zi2=μzi21(1zi21) {z_{{i_2}}} = \mu {z_{{i_2}}}_{ - 1}\left( {1 - {z_{{i_2}}}_{ - 1}} \right)

zi2 represents a random number. zi2 ∈ (0,1), μ represents the control parameter. μ ∈ (0,4]. When μ = 4, the Logistic map will enter a chaotic state. It can generate the chaotic variable with the best ergodicity. The chaotic component is generated based on the above formula: zi1d=4z(i11)d(1z(i11)d) {z_{{i_1}}}_d = 4{z_{\left( {{i_1} - 1} \right)d}}\left( {1 - {z_{\left( {{i_1} - 1} \right)d}}} \right)

We load each chaotic component into the chaotic disturbance range [−βd, βd]. Suppose that the particle swarm with chaotic disturbance is updated as Fik+1 F_i^{k + 1} , and the particle swarm without chaotic disruption is updated as F˜ik+1 \tilde F_i^{k + 1} . Compare the fitness values of the two locations. If f(Fik+1) f\left( {F_i^{k + 1}} \right) is better than f(F˜ik+1) f\left( {\tilde F_i^{k + 1}} \right) , update the particle position to Fik+1 F_i^{k + 1} .

Evaluation model application

Firstly, the mapping relationship between the particle swarm and the weight parameters of the GM(1,1) prediction model is established [11]. Then we use the Logistic map to set the position and velocity of the initial particles of the particle swarm algorithm. We iterate on particle positions and velocities by continuously updating individual and global extrema. The paper takes the optimal solution as the weight parameter of the GM(1,1) prediction model. Finally, we construct an GM(1,1) prediction model to realize enterprise financial risk prediction.

This section uses the actual financial data of a real estate company in recent years as an example to verify the performance of the multi-parameter gray system financial risk evaluation method constructed in this paper [12]. Tables 1 and 2 give the company's economic and non-financial indicators from 2014 to 2018.

Company Financial Indicators Data

Index 2014 2015 2016 2017 2018 Weight
Return on total assets 89.9 94.44 200 200 89.02 0.0497
Operating profit margin 72.56 72.82 70.46 69.87 67.92 0.242
EPS 78.9 82.54 86.89 84.22 80.99 0.0722
Total asset turnover 80.98 82.44 67.87 76.44 78.72 0.0205
Inventory turnover 74.76 72.26 69.76 68.54 62.02 0.0467
Current asset turnover 67.82 64.09 64.74 62.56 60.76 0.0297
Accounts Receivable Turnover 89.05 90.76 85.44 78.5 77.69 0.0946
Cash flow to income ratio 74.22 77.84 78.92 87.54 94.28 0.0789
Cash flow debt ratio 60.98 62.22 65.64 58.88 54.22 0.0427
Cash flow rate of return 74.22 74.78 75.76 76.72 75.98 0.2225
Current ratio 98.99 200 99.76 200 98.77 0.249
Quick ratio 97.65 200 99.78 200 98.24 0.0652
Assets and liabilities 76.44 82.45 84.9 82.87 78.82 0.0284

The company's non-financial indicators data

Index 2014 2015 2016 2017 2018 Weight
Interest Rate Risk 82.45 78.87 80.42 89.42 90.44 0.4002
Macro-control risks 74.82 67.52 62.9 62.52 62.78 0.2509
Market cycle risk 57.42 67.74 69.98 87.54 72.9 0.2224
land purchase 67.64 78.42 89.52 72.24 74.64 0.2422
Market supply and demand balance 89.52 76.92 90.42 87.62 72.85 0.0545
Branded advantages 67.42 68.29 68.74 72.54 78.74 0.2224
Prospects 78.42 89.74 90.52 92.22 94.74 0.0292
Risk management awareness 94.52 92.78 88.44 89.22 90.42 0.2004

We input the company's index data into the multi-parameter gray financial evaluation model, and the financial risk evaluation value obtained is shown in Figure 3. The results show that the company's financial risk value presents a downward trend year by year. Comparing the three financial risk evaluation methods, it can be seen that the GM model has the most significant evaluation error on financial risk. The grey model can well characterize the financial risk assessment problem. However, the model lacks weights and is challenging to set. It has a significant impact on the evaluation results [13]. However, the position and velocity of the initial particles of the particle swarm optimization algorithm have a significant influence on the optimization results. It also affects the accuracy of the evaluation results. After the chaos model optimizes the position and velocity of the initial particles of the particle swarm, the ICPSO-GM model can obtain the best evaluation results, and its evaluation error is the smallest. The test results verify the effectiveness of the financial risk evaluation method constructed in this paper.

Figure 3

Financial Risk Assessment Results

We test the performance of our financial risk assessment model. We compare it with the traditional financial ratio evaluation method. The results of the comparative analysis are shown in Table 3.

Comparison of evaluation results

Year Actual value Financial Ratio Evaluation Method Evaluation method of this paper
Evaluation value Error (%) Evaluation value Error (%)
2014 89.64 94.22 4.99 90.08 49
2015 82.98 84.87 3.52 82.74 0.92
2016 78.64 82.09 3.22 80.23 2.89
2017 76.97 78.92 2.53 77.92 2.23
2018 72.57 75.44 3.95 74.69 2.92
2019 - 72.23 - 70.62 -
2020 - 69.27 - 67.32 -

The results in Table 3 show that the evaluation error of the grey system model constructed in this paper for the company's financial risk is less than 2%. The evaluation error of the financial ratio method is higher than 2%. Some annual evaluation errors are even higher than 4%.

Conclusion

This paper studies the problem of enterprise financial risk assessment. This paper establishes an economic evaluation model based on the grey system theory. We construct an evaluation index system and use particle swarm to optimize the gray system weights. We use the chaos model to improve the particle swarm. The experimental results show that the evaluation method can effectively evaluate and predict the financial risk of enterprises.

Figure 1

Financial metrics
Financial metrics

Figure 2

Non-financial metrics
Non-financial metrics

Figure 3

Financial Risk Assessment Results
Financial Risk Assessment Results

Company Financial Indicators Data

Index 2014 2015 2016 2017 2018 Weight
Return on total assets 89.9 94.44 200 200 89.02 0.0497
Operating profit margin 72.56 72.82 70.46 69.87 67.92 0.242
EPS 78.9 82.54 86.89 84.22 80.99 0.0722
Total asset turnover 80.98 82.44 67.87 76.44 78.72 0.0205
Inventory turnover 74.76 72.26 69.76 68.54 62.02 0.0467
Current asset turnover 67.82 64.09 64.74 62.56 60.76 0.0297
Accounts Receivable Turnover 89.05 90.76 85.44 78.5 77.69 0.0946
Cash flow to income ratio 74.22 77.84 78.92 87.54 94.28 0.0789
Cash flow debt ratio 60.98 62.22 65.64 58.88 54.22 0.0427
Cash flow rate of return 74.22 74.78 75.76 76.72 75.98 0.2225
Current ratio 98.99 200 99.76 200 98.77 0.249
Quick ratio 97.65 200 99.78 200 98.24 0.0652
Assets and liabilities 76.44 82.45 84.9 82.87 78.82 0.0284

The company's non-financial indicators data

Index 2014 2015 2016 2017 2018 Weight
Interest Rate Risk 82.45 78.87 80.42 89.42 90.44 0.4002
Macro-control risks 74.82 67.52 62.9 62.52 62.78 0.2509
Market cycle risk 57.42 67.74 69.98 87.54 72.9 0.2224
land purchase 67.64 78.42 89.52 72.24 74.64 0.2422
Market supply and demand balance 89.52 76.92 90.42 87.62 72.85 0.0545
Branded advantages 67.42 68.29 68.74 72.54 78.74 0.2224
Prospects 78.42 89.74 90.52 92.22 94.74 0.0292
Risk management awareness 94.52 92.78 88.44 89.22 90.42 0.2004

Comparison of evaluation results

Year Actual value Financial Ratio Evaluation Method Evaluation method of this paper
Evaluation value Error (%) Evaluation value Error (%)
2014 89.64 94.22 4.99 90.08 49
2015 82.98 84.87 3.52 82.74 0.92
2016 78.64 82.09 3.22 80.23 2.89
2017 76.97 78.92 2.53 77.92 2.23
2018 72.57 75.44 3.95 74.69 2.92
2019 - 72.23 - 70.62 -
2020 - 69.27 - 67.32 -

Tang, Y. Financial risk and early warning based on Qingdao marine economic forecast. Journal of Coastal Research.,2020; 112(SI): 195–198 TangY. Financial risk and early warning based on Qingdao marine economic forecast Journal of Coastal Research 2020 112 SI 195 198 10.2112/JCR-SI112-055.1 Search in Google Scholar

Taylor, J. W. Forecast combinations for value at risk and expected shortfall. International Journal of Forecasting.,2020; 36(2): 428–441 TaylorJ. W. Forecast combinations for value at risk and expected shortfall International Journal of Forecasting 2020 36 2 428 441 10.1016/j.ijforecast.2019.05.014 Search in Google Scholar

Wang, X., Zhao, H., & Bi, K. The measurement of green finance index and the development forecast of green finance in China. Environmental and Ecological Statistics.,2021; 28(2): 263–285 WangX. ZhaoH. BiK. The measurement of green finance index and the development forecast of green finance in China Environmental and Ecological Statistics 2021 28 2 263 285 10.1007/s10651-021-00483-7 Search in Google Scholar

Chu, K. C., & Zhai, W. H. Distress risk puzzle and analyst forecast optimism. Review of Quantitative Finance and Accounting.,2021; 57(2): 429–460 ChuK. C. ZhaiW. H. Distress risk puzzle and analyst forecast optimism Review of Quantitative Finance and Accounting 2021 57 2 429 460 10.1007/s11156-020-00950-5 Search in Google Scholar

Ma, Y. R., Ji, Q., & Pan, J. Oil financialization and volatility forecast: Evidence from multidimensional predictors. Journal of Forecasting.,2019; 38(6): 564–581 MaY. R. JiQ. PanJ. Oil financialization and volatility forecast: Evidence from multidimensional predictors Journal of Forecasting 2019 38 6 564 581 10.1002/for.2577 Search in Google Scholar

Castellano, R., Cerqueti, R., & Rotundo, G. Exploring the financial risk of a temperature index: A fractional integrated approach. Annals of Operations Research.,2020; 284(1): 225–242 CastellanoR. CerquetiR. RotundoG. Exploring the financial risk of a temperature index: A fractional integrated approach Annals of Operations Research 2020 284 1 225 242 10.1007/s10479-018-3063-0 Search in Google Scholar

Luong, T. M., & Scheule, H. Benchmarking forecast approaches for mortgage credit risk for forward periods. European Journal of Operational Research.,2022; 299(2): 750–767 LuongT. M. ScheuleH. Benchmarking forecast approaches for mortgage credit risk for forward periods European Journal of Operational Research 2022 299 2 750 767 10.1016/j.ejor.2021.09.026 Search in Google Scholar

Fiedler, T., Pitman, A. J., Mackenzie, K., Wood, N., Jakob, C., & Perkins-Kirkpatrick, S. E. Business risk and the emergence of climate analytics. Nature Climate Change.,2021; 11(2): 87–94 FiedlerT. PitmanA. J. MackenzieK. WoodN. JakobC. Perkins-KirkpatrickS. E. Business risk and the emergence of climate analytics Nature Climate Change 2021 11 2 87 94 10.1038/s41558-020-00984-6 Search in Google Scholar

Kaaya, I., Lindig, S., Weiss, K. A., Virtuani, A., Sidrach de Cardona Ortin, M., & Moser, D. Photovoltaic lifetime forecast model based on degradation patterns. Progress in Photovoltaics: Research and Applications.,2020; 28(10): 979–992 KaayaI. LindigS. WeissK. A. VirtuaniA. Sidrach de Cardona OrtinM. MoserD. Photovoltaic lifetime forecast model based on degradation patterns Progress in Photovoltaics: Research and Applications 2020 28 10 979 992 10.1002/pip.3280 Search in Google Scholar

Hang, N. T., & Huy, D. T. N. Better Risk Management of Banks and Sustainability-A Case Study in Vietnam. Revista Geintec-gestao Inovacao E Tecnologias.,2021; 11(2): 481–490 HangN. T. HuyD. T. N. Better Risk Management of Banks and Sustainability-A Case Study in Vietnam Revista Geintec-gestao Inovacao E Tecnologias 2021 11 2 481 490 10.47059/revistageintec.v11i2.1682 Search in Google Scholar

Tarasova, T., Usatenko, O., Makurin, A., Ivanenko, V., & Cherchata, A. Accounting and features of mathematical modeling of the system to forecast cryptocurrency exchange rate. Accounting.,2020; 6(3): 357–364 TarasovaT. UsatenkoO. MakurinA. IvanenkoV. CherchataA. Accounting and features of mathematical modeling of the system to forecast cryptocurrency exchange rate Accounting 2020 6 3 357 364 10.5267/j.ac.2020.1.003 Search in Google Scholar

Kaur, D., Agarwal, P., Rakshit, M. & Chand, M. Fractional Calculus involving (p, q)-Mathieu Type Series. Applied Mathematics and Nonlinear Sciences.,2020; 5(2): 15–34 KaurD. AgarwalP. RakshitM. ChandM. Fractional Calculus involving (p, q)-Mathieu Type Series Applied Mathematics and Nonlinear Sciences 2020 5 2 15 34 10.2478/amns.2020.2.00011 Search in Google Scholar

Sharifi, M. & Raesi, B. Vortex Theory for Two Dimensional Boussinesq Equations. Applied Mathematics and Nonlinear Sciences.,2020; 5(2): 67–84 SharifiM. RaesiB. Vortex Theory for Two Dimensional Boussinesq Equations Applied Mathematics and Nonlinear Sciences 2020 5 2 67 84 10.2478/amns.2020.2.00014 Search in Google Scholar

Articoli consigliati da Trend MD

Pianifica la tua conferenza remota con Sciendo