The rapid development of financial technology not only provides a lot of convenience to people's production and life, but also brings a lot of risks to financial security. To prevent financial risks, a better way is to build an accurate warning model before the financial risk occurs, not to find a solution after the outbreak of the risk. In the past decade, deep learning has made amazing achievements in the fields, such as image recognition, natural language processing. Therefore, some researchers try to apply deep learning methods to financial risk prediction and most of the results are satisfactory. The main work of this paper is to review the predecessors' work of deep learning for financial risk prediction according to three prominent characteristics of financial data: heterogeneity, multi-source, and imbalance. We first briefly introduced some classical deep learning models as the model basis of financial risk prediction. Then we analyzed the reasons for these characteristics of financial data. Meanwhile, we studied the differences of commonly used deep learning models according to different data characteristics. Finally, we pointed out some open issues with research significance in this field and suggested the future implementations that might be feasible.
Citation: Kuashuai Peng, Guofeng Yan. A survey on deep learning for financial risk prediction[J]. Quantitative Finance and Economics, 2021, 5(4): 716-737. doi: 10.3934/QFE.2021032
[1] | Bilal Akbar, Khuram Pervez Amber, Anila Kousar, Muhammad Waqar Aslam, Muhammad Anser Bashir, Muhammad Sajid Khan . Data-driven predictive models for daily electricity consumption of academic buildings. AIMS Energy, 2020, 8(5): 783-801. doi: 10.3934/energy.2020.5.783 |
[2] | Zhongjiao Ma, Zichun Yan, Mingfei He, Haikuan Zhao, Jialin Song . A review of the influencing factors of building energy consumption and the prediction and optimization of energy consumption. AIMS Energy, 2025, 13(1): 35-85. doi: 10.3934/energy.2025003 |
[3] | Kaovinath Appalasamy, R Mamat, Sudhakar Kumarasamy . Smart thermal management of photovoltaic systems: Innovative strategies. AIMS Energy, 2025, 13(2): 309-353. doi: 10.3934/energy.2025013 |
[4] | Feng Tian . A wind power prediction model with meteorological time delay and power characteristic. AIMS Energy, 2025, 13(3): 517-539. doi: 10.3934/energy.2025020 |
[5] | Xiaojing Tian, Weiqi Ye, Liang Xu, Anjian Yang, Langming Huang, Shenglong Jin . Optimization research on laminated cooling structure for gas turbines: A review. AIMS Energy, 2025, 13(2): 354-401. doi: 10.3934/energy.2025014 |
[6] | Emrah Gulay . Forecasting the total electricity production in South Africa: Comparative analysis to improve the predictive modelling accuracy. AIMS Energy, 2019, 7(1): 88-110. doi: 10.3934/energy.2019.1.88 |
[7] | Rahaman Abu, John Amakor, Rasaq Kazeem, Temilola Olugasa, Olusegun Ajide, Nosa Idusuyi, Tien-Chien Jen, Esther Akinlabi . Modeling influence of weather variables on energy consumption in an agricultural research institute in Ibadan, Nigeria. AIMS Energy, 2024, 12(1): 256-270. doi: 10.3934/energy.2024012 |
[8] | Yan Li, Yaheng Su, Qixin Zhao, Bala Wuda, Kaibo Qu, Lei Tang . An electricity price optimization model considering time-of-use and active distribution network efficiency improvements. AIMS Energy, 2025, 13(1): 13-34. doi: 10.3934/energy.2025002 |
[9] | Anbazhagan Geetha, S. Usha, J. Santhakumar, Surender Reddy Salkuti . Forecasting of energy consumption rate and battery stress under real-world traffic conditions using ANN model with different learning algorithms. AIMS Energy, 2025, 13(1): 125-146. doi: 10.3934/energy.2025005 |
[10] | Ghafi Kondi Akara, Benoit Hingray, Adama Diawara, Arona Diedhiou . Effect of weather on monthly electricity consumption in three coastal cities in West Africa. AIMS Energy, 2021, 9(3): 446-464. doi: 10.3934/energy.2021022 |
The rapid development of financial technology not only provides a lot of convenience to people's production and life, but also brings a lot of risks to financial security. To prevent financial risks, a better way is to build an accurate warning model before the financial risk occurs, not to find a solution after the outbreak of the risk. In the past decade, deep learning has made amazing achievements in the fields, such as image recognition, natural language processing. Therefore, some researchers try to apply deep learning methods to financial risk prediction and most of the results are satisfactory. The main work of this paper is to review the predecessors' work of deep learning for financial risk prediction according to three prominent characteristics of financial data: heterogeneity, multi-source, and imbalance. We first briefly introduced some classical deep learning models as the model basis of financial risk prediction. Then we analyzed the reasons for these characteristics of financial data. Meanwhile, we studied the differences of commonly used deep learning models according to different data characteristics. Finally, we pointed out some open issues with research significance in this field and suggested the future implementations that might be feasible.
In the present paper, we study the following coupled nonlinear wave equations (see [1])
utt−div(ρ(|∇u|2)∇u)+|ut|m−1ut+f1(u,v)=0, | (1.1) |
vtt−div(ρ(|∇v|2)∇v)+|vt|n−1vt+f2(u,v)=0, | (1.2) |
where ut=∂u∂t,vt=∂v∂t, m,n>1, ∇ is the gradient operator, div is the divergence operator and fi(.,.):R2→R,i=1,2 are known functions. For arbitrary solutions of (1.1) and (1.2), the function ρ is supposed to satisfy one or the other of the two conditions:
A1). 0<ρ2(q2)q2≤m1˜ρ(q2),
or
A2). 0<1K1<ρ−1(q2)≤K2˜ρ(q2)1q2,
where ˜ρ(q2)=∫q20ρ(ζ)dζ, q2=|∇u|2,m1,K1,K2>0.
Quintanilla [2] imposed condition A1 and obtained the growth or decay estimates of the solution to the type Ⅲ heat conduction. The condition A1 can be satisfied easily, e.g., ρ(q2)=1√1+q2 or ρ(q2)=(1+bpq2)p−1,b>0,0<p≤1. The similar condition as A2 was also considered by many papers (see [3,4]). ρ(q2)=√1+q2 satisfies the condition A2.
In addition, we introduce a function F(u,v) which is defined as ∂F∂u=f1(u,v), ∂F∂v=f2(u,v), where F(0,0)=0.
The wave equations have attracted many attentions of scholars due to their wide application, and a large number of achievements have been made in the existence of solutions (see [1,5,6,7,8,9,10,11,12]). The fuzzy inference method is used to solve this problem. The algebraic formulation of fuzzy relation is studied in [13,14]. In this paper, we study the Phragmén-Lindelöf type alternative property of solutions of wave equations (1.1)–(1.2). It is proved that the solution of the equations either grows exponentially (polynomially) or decays exponentially (polynomially) when the space variable tends to infinite. In the case of decay, people usually expect a fast decay rate. The Phragmén-Lindelöf type alternative research on partial differential equations has lasted for a long time (see [2,15,16,17,18,19,20,21,22,23]).
It is worth emphasizing that Quintanilla [2] considered an exterior or cone-like region. Under some appropriate conditions, the growth/decay estimates of some parabolic problems are obtained. Inspired by [2], we extends the research of to the nonlinear wave model in this paper. However, different from [2], in addition to condition A1 and condition A2, we also consider a special condition of ρ. The appropriate energy function is established, and the nonlinear differential inequality about the energy function is derived. By solving this differential inequality, the Phragmén-Lindelöf type alternative results of the solution are obtained. Our model is much more complex than [2], so the methods used in [2] can not be applied to our model directly. Finally, a nonlinear system of viscoelastic type is also studied when the system is defined in an exterior or cone-like regions and the growth or decay rates are also obtained.
Letting that Ω(r) denotes a cone-like region, i.e.,
Ω(r)={x||x|2≥r2,r≥R0>0}, |
and that B(r) denotes the exterior surface to the sphere, i.e.,
B(r)={x||x|2=r2,r≥R0>0}, |
Equations (1.1) and (1.2) also have the following initial-boundary conditions
u(x,0)=v(x,0)=0, in Ω, | (2.1) |
u(x,t)=g1(x,t), v(x,t)=g2(x,t), in B(R0)×(0,τ), | (2.2) |
where g1 and g2 are positive known functions, x=(x1,x2,x3) and τ>0.
Now, we establish an energy function
E(r,t)=∫t0∫B(r)e−ωηρ(|∇u|2)∇u⋅xruηdSdη+∫t0∫B(r)e−ωηρ(|∇v|2)∇v⋅xrvηdSdη≐E1(r,t)+E2(r,t). | (2.3) |
Let r0 be a positive constant which satisfies r>r0≥R0. Integrating E(z,t) from r0 to r, using the divergence theorem and Eqs (1.1) and (1.2), (2.1) and (2.2), we have
E(r,t)−E(r0,t)=∫t0∫rr0∫B(ξ)e−ωη∇⋅[ρ(|∇u|2)∇uuη]dsdξdη+∫t0∫rr0∫B(ξ)e−ωη∇⋅[ρ(|∇v|2)∇vvη]dsdξdη+∫t0∫rr0∫B(ξ)e−ωη[uηη+|uη|m+1+f1(u,v)]uηdsdξdη+∫t0∫rr0∫B(ξ)e−ωη[vηη+|vη|n+1+f2(u,v)]vηdsdξdη+12∫t0∫rr0∫B(ξ)e−ωη∂∂η˜ρ(|∇u|2)dsdξdη+12∫t0∫rr0∫B(ξ)e−ωη∂∂η˜ρ(|∇v|2)dsdξdη=12e−ωt∫rr0∫B(ξ)[|ut|2+|vt|2+˜ρ(|∇u|2)+˜ρ(|∇v|2)+2F(u,v)]dsdξ+12ω∫t0∫rr0∫B(ξ)e−ωη[|uη|2+|vη|2+˜ρ(|∇u|2)+˜ρ(|∇v|2)+2F(u,v)]dSdξdη+∫t0∫rr0∫B(ξ)e−ωη[|uη|m+1+|vη|n+1]dsdξdη, | (2.4) |
from which it follows that
∂∂rE(r,t)=12e−ωt∫B(r)[|ut|2+|vt|2+˜ρ(|∇u|2)+˜ρ(|∇v|2)+2F(u,v)]ds+12ω∫t0∫B(r)e−ωη[|uη|2+|vη|2+˜ρ(|∇u|2)+˜ρ(|∇v|2)+2F(u,v)]dSdη+∫t0∫B(r)e−ωη[|uη|m+1+|vη|n+1]dsdη, | (2.5) |
where ω is positive constant.
Now, we show how to bound E(r,t) by ∂∂rE(r,t). We use the Hölder inequality, the Young inequality and the condition A1 to have
|E1(r,t)|≤[∫t0∫B(r)e−ωηρ2(|∇u|2)|∇u|2dsdη⋅∫t0∫B(r)e−ωηu2ηdsdη]12≤√m1[∫t0∫B(r)e−ωη˜ρ(|∇u|2)dsdη⋅∫t0∫B(r)e−ωηu2ηdsdη]12≤√m12[∫t0∫B(r)e−ωη˜ρ(|∇u|2)dsdη+∫t0∫B(r)e−ωηu2ηdsdη], | (2.6) |
and
|E2(r,t)|≤√m12[∫t0∫B(r)e−ωη˜ρ(|∇v|2)dsdη+∫t0∫B(r)e−ωηv2ηdsdη]. | (2.7) |
Inserting (2.6) and (2.7) into (2.3) and combining (2.5), we have
|E(r,t)|≤√m1ω[∂∂rE(r,t)]. | (2.8) |
We consider inequality (2.8) for two cases.
I. If ∃r0>R0 such that E(r0,t)≥0. From (2.5), we obtain E(r,t)≥E(r0,t)≥0,r≥r0. Therefore, from (2.8) we have
E(r,t)≤√m1ω[∂∂rE(r,t)],r≥r0. | (2.9) |
Integrating (2.9) from r0 to r, we have
E(r,t)≥[E(r0,t)]eω√m1(r−r0),r≥r0. | (2.10) |
Combing (2.4) and (2.10), we have
limr→∞{e−ω√m1r[12e−ωt∫rr0∫B(ξ)[|ut|2+|vt|2+˜ρ(|∇u|2)+˜ρ(|∇v|2)+2F(u,v)]dsdξ+12ω∫t0∫rr0∫B(ξ)e−ωη[|uη|2+|vη|2+˜ρ(|∇u|2)+˜ρ(|∇v|2)+2F(u,v)]dsdξdη+∫t0∫rr0∫B(ξ)e−ωη[|uη|m+1+|vη|n+1]dsdξdη]}≥E(R0,t)e−ω√m1R0. | (2.11) |
II. If ∀r>R0 such that E(r,t)<0. The inequality (2.8) can be rewritten as
−E(r,t)≤√m1ω[∂∂rE(r,t)],r≥R0. | (2.12) |
Integrating (2.12) from r0 to r, we have
−E(r,t)≥[−E(R0,t)]e−ω√m1(r−R0),r≥R0. | (2.13) |
Inequality (2.13) shows that limr→∞[−E(r,t)]=0. Integrating (2.5) from r to ∞ and combining (2.13), we obtain
12e−ωt∫∞r∫B(ξ)[|ut|2+|vt|2+˜ρ(|∇u|2)+˜ρ(|∇v|2)+2F(u,v)]dsdξ+12ω∫t0∫∞r∫B(ξ)e−ωη[|uη|2+|vη|2+˜ρ(|∇u|2)+˜ρ(|∇v|2)+2F(u,v)]dsdξdη+∫t0∫∞r∫B(ξ)e−ωη[|uη|m+1+|vη|n+1]dsdξdη≤[−E(r0,t)]e−ω√m1(r−R0). | (2.14) |
We summarize the above result as the following theorem.
Theorem 2.1. Let (u,v) be solution of the (1.1), (1.2), (2.1), (2.2) in Ω(R0), and ρ satisfies condition A1. Then for fixed t, (u,v) either grows exponentially or decays exponentially. Specifically, either (2.11) holds or (2.14) holds.
If the function ρ satisfies the condition A2, we recompute the inequality (2.6) and (2.7). Therefore
|E1(r,t)|≤[∫t0∫B(r)e−ωηρ2(|∇u|2)|∇u|2dsdη⋅∫t0∫B(r)e−ωηu2ηdsdη]12≤K1√K1[∫t0∫B(r)e−ωηρ−1(|∇u|2)dsdη⋅∫t0∫B(r)e−ωηu2ηdsdη]12≤K1√K1K22[∫t0∫B(r)e−ωη˜ρ(|∇u|2)dsdη+∫t0∫B(r)e−ωηu2ηdsdη], | (3.1) |
and
|E2(r,t)|≤K1√K1K22[∫t0∫B(r)e−ωη˜ρ(|∇v|2)dsdη+∫t0∫B(r)e−ωηv2ηdsdη]. | (3.2) |
Inserting (3.1) and (3.2) into (2.3) and combining (2.5), we have
|E(r,t)|≤K1√K1K2ω[∂∂rE(r,t)]. | (3.3) |
By following a similar method to that used in Section 2, we can obtain the Phragmén-Lindelöf type alternative result.
Theorem 3.1. Let (u,v) be solution of the (1.1), (1.2), (2.1), (2.2) in Ω(R0), and ρ satisfies condition A1. Then for fixed t, (u,v) either grows exponentially or decays exponentially.
Remark 3.1. Clearly, the rates of growth or decay obtained in Theorems 2.1 and 3.1 depend on ω. Because ω can be chosen large enough, the rates of growth or decay of the solutions can become large as we want.
Remark 3.2. The analysis in Sections 2 and 3 can be adapted to the single-wave equation
utt−div(ρ(|∇u|2)∇u)+|ut|m−1ut+f(u)=0 | (3.4) |
and the heat conduction at low temperature
autt+but−cΔu+Δut=0, | (3.5) |
where a,b,c>0.
In this section, we suppose that ρ satisfies ρ(q2)=b1+b2q2β, where b1,b2 and β are positive constants. Clearly, ρ(q2)=b1+b2q2β can not satisfy A1 or A2. In this case, we define an "energy" function
F(r,t)=∫t0∫B(r)e−ωη(b1+b2|∇u|2β)∇u⋅xruηdsdη+∫t0∫B(r)e−ωη(b1+b2|∇v|2β)∇v⋅xrvηdsdη≐F1(r,t)+F2(r,t). | (4.1) |
Computing as that in (2.4) and (2.5), we can get
F(r,t)=F(r0,t)+12e−ωt∫rr0∫B(ξ)[|ut|2+b1|∇u|2+1β+1b2|∇u|2(β+1)]dsdξ+12e−ωt∫rr0∫B(ξ)[|vt|2+b1|∇v|2+1β+1b2|∇v|2(β+1)+F(u,v)]dsdξ+12ω∫t0∫rr0∫B(ξ)e−ωη[|uη|2+b1|∇u|2+1β+1b2|∇u|2(β+1)]dsdξdη+12ω∫t0∫rr0∫B(ξ)e−ωη[|vη|2+b1|∇v|2+1β+1b2|∇v|2(β+1)+F(u,v)]dsdξdη+∫t0∫rr0∫B(ξ)e−ωη|uη|m+1dsdξdη+∫t0∫B(r)e−ωη|vη|n+1dsdξdη, | (4.2) |
and
∂∂rF(r,t)=12e−ωt∫B(r)[|ut|2+b1|∇u|2+1β+1b2|∇u|2(β+1)]ds+12e−ωt∫B(r)[|vt|2+b1|∇v|2+1β+1b2|∇v|2(β+1)+F(u,v)]ds+12ω∫t0∫B(r)e−ωη[|uη|2+b1|∇u|2+1β+1b2|∇u|2(β+1)]dsdη+12ω∫t0∫B(r)e−ωη[|vη|2+b1|∇v|2+1β+1b2|∇v|2(β+1)+F(u,v)]dsdη+∫t0∫B(r)e−ωη|uη|m+1dsdη+∫t0∫B(r)e−ωη|vη|n+1dsdη. | (4.3) |
Using the Hölder inequality and Young's inequality, we have
|F1(r,t)|≤b1[∫t0∫B(r)e−ωη|∇u|2dsdη∫t0∫B(r)e−ωηu2ηdsdη]12+b2[∫t0∫B(r)e−ωη|∇u|2(β+1)dsdη]2β+12(β+1)∫t0∫B(r)e−ωη|uη|m+1dsdη]1m+1|t2πr|12(β+1)−1m+1≤√b1R0ω[b12rω∫t0∫B(r)e−ωη|∇u|2dsdη+12rω∫t0∫B(r)e−ωηu2ηdsdη]+b2|2tπ|12(β+1)−1m+1R−ββ+1−2m+10[2β+12(β+1)r2β+12(β+1)+1m+1∫t0∫B(r)e−ωη|∇u|2(β+1)dsdη+1m+1r2β+12(β+1)+1m+1∫t0∫B(r)e−ωη|uη|m+1dsdη]2β+12(β+1)+1m+1, | (4.4) |
and
|F2(r,t)|≤√b1R0ω[b12rω∫t0∫B(r)e−ωη|∇v|2dsdη+12rω∫t0∫B(r)e−ωηv2ηdsdη]+b2|2tπ|12(β+1)−1n+1R−ββ+1−2n+10[2β+12(β+1)r2β+12(β+1)+1n+1∫t0∫B(r)e−ωη|∇v|2(β+1)dsdη+1n+1r2β+12(β+1)+1n+1∫t0∫B(r)e−ωη|vη|n+1dsdη]2β+12(β+1)+1n+1, | (4.5) |
where we have chosen 12(β+1)>1n+1. Inserting (4.4) and (4.5) into (4.1), we obtain
|F(r,t)|≤c1[r∂∂rF(r,t)]+c2[r∂∂rF(r,t)]2β+12(β+1)+1m+1+c3[r∂∂rF(r,t)]2β+12(β+1)+1n+1, | (4.6) |
where c1=√b1R0ω,c2=b2|2tπ|12(β+1)−1n+1R−ββ+1−2m+10,c3=b2|2tπ|12(β+1)−1n+1R−ββ+1−2n+10.
Next, we will analyze Eq (4.6) in two cases
I. If ∃r0≥R0 such that F(r0,t)≥0, then F(r,t)≥F(r0,t)≥0,r≥r0. Therefore, (4.6) can be rewritten as
F(r,t)≤c1[r∂∂rF(r,t)]+c2[r∂∂rF(r,t)]2β+12(β+1)+1m+1+c3[r∂∂rF(r,t)]2β+12(β+1)+1n+1,r≥r0. | (4.7) |
Using Young's inequality, we have
[r∂∂rF(r,t)]2β+12(β+1)+1m+1≤(4β+34(β+1)+12(m+1))[r∂∂rF(r,t)]12+(14(β+1)−12(m+1))[r∂∂rF(r,t)], | (4.8) |
[r∂∂rF(r,t)]2β+12(β+1)+1n+1≤(4β+34(β+1)+12(n+1))[r∂∂rF(r,t)]12+(14(β+1)−12(n+1))[r∂∂rF(r,t)]. | (4.9) |
Inserting (4.8) and (4.9) into (4.7), we have
F(r,t)≤c4[r∂∂rF(r,t)]12+c5[r∂∂rF(r,t)],r≥r0. | (4.10) |
where c4=c2(4β+34(β+1)+12(m+1))+c3(4β+34(β+1)+12(n+1)) and c5=c1+c2(14(β+1)−12(m+1))+c3(14(β+1)−12(n+1)). From (4.10) we have
F(r,t)c5≤[√r∂∂rF(r,t)+c42c5]2−c244c25,r≥r0 |
or
∂∂rF(r,t)[√F(r,t)c5+c244c25−c42c5]2≥1r,r≥r0 | (4.11) |
Integrating (4.11) from r0 to r, we get
2c5[ln(√F(r,t)c5+c244c25−c42c5)−ln(√F(r0,t)c5+c244c25−c42c5)]−c4{[√F(r,t)c5+c244c25−c42c5]−1−[√F(r0,t)c5+c244c25−c42c5]−1}≥ln(rr0),r≥r0 | (4.12) |
Dropping the third term on the left of (4.12), we have
√F(r,t)c5+c244c25−c42c5≥Q1(r0,t)(rr0)12c5, | (4.13) |
where Q1(r0,t)=(√F(r0,t)c5+c244c25−c42c5)exp{−c42c5[√F(r0,t)c5+c244c25−c42c5]−1}.
In view of
√F(r,t)c5+c244c25≤√F(r,t)c5+c42c5, |
we have from (4.13)
F(r,t)≥c5Q21(r0,t)(rr0)1c5. | (4.14) |
Combining (4.2) and (4.14), we have
limr→∞{r−1c5[12e−ωt∫rr0∫B(ξ)[|ut|2+b1|∇u|2+1β+1b2|∇u|2(β+1)]dsdξ+12e−ωt∫rr0∫B(ξ)[|vt|2+b1|∇v|2+1β+1b2|∇v|2(β+1)+F(u,v)]dsdξ+12ω∫t0∫rr0∫B(ξ)e−ωη[|uη|2+b1|∇u|2+1β+1b2|∇u|2(β+1)]dsdξdη+12ω∫t0∫rr0∫B(ξ)e−ωη[|vη|2+b1|∇v|2+1β+1b2|∇v|2(β+1)+F(u,v)]dsdξdη+∫t0∫rr0∫B(ξ)e−ωη|uη|m+1dsdξdη+∫t0∫B(r)e−ωη|vη|n+1dsdξdη]}≥c5Q21(r0,t)r−1c50. | (4.15) |
II. If ∀r≥R0 such that F(r,t)<0, then (4.6) can be rewritten as
−F(r,t)≤c1[r∂∂rF(r,t)]+c2[r∂∂rF(r,t)]2β+12(β+1)+1m+1+c3[r∂∂rF(r,t)]2β+12(β+1)+1n+1,r≥R0. | (4.16) |
Without losing generality, we suppose that m>n>1.
[r∂∂rF(r,t)]2β+12(β+1)+1n+1≤1n+1−1m+12β+12(β+1)−1m+1[r∂∂rF(r,t)]+2β+12(β+1)−1m+12β+12(β+1)−1m+1[r∂∂rF(r,t)]2β+12(β+1)+1m+1, | (4.17) |
[r∂∂rF(r,t)]≤2β(m+3)+4(2β+1)(m+2)+1[r∂∂rF(r,t)]2β+12(β+1)+1m+1+m+1−2(β+1)(2β+1)(m+2)+1[r∂∂rF(r,t)]2[2β+12(β+1)+1m+1]. | (4.18) |
Inserting (4.17) and (4.18) into (4.16), we get
−F(r,t)≤c6[r∂∂rF(r,t)]2β+12(β+1)+1m+1+c7[r∂∂rF(r,t)]2[2β+12(β+1)+1m+1], r≥R0, | (4.19) |
where c6=[c1+c31n+1−1m+12β+12(β+1)−1m+1]2β(m+3)+4(2β+1)(m+2)+1+c32β+12(β+1)−1m+12β+12(β+1)−1m+1,c7=c3m+1−2(β+1)(2β+1)(m+2)+1. From (4.19) we obtain
r∂∂rF(r,t)≥[√−F(r,t)c7+c264c27−c62c7]12β+12(β+1)+1m+1, r≥R0 |
or
2c7√−F(r,t)c7+c264c27[√−F(r,t)c7+c264c27−c62c7]12β+12(β+1)+1m+1d{√−F(r,t)c7+c264c27−c62c7}≤−1r, r≥R0 | (4.20) |
Integrating (4.20) from R0 to r, we obtain
2c7(2β+12(β+1)+1m+1)β2(β+1)+2m+1[√−F(r,t)c7+c264c27−c62c7]β2(β+1)+2m+12β+12(β+1)+1m+1−2c7(2β+12(β+1)+1m+1)β2(β+1)+2m+1[√−F(R0,t)c7+c264c27−c62c7]β2(β+1)+2m+12β+12(β+1)+1m+1−c6(2β+12(β+1)+1m+1)12(β+1)−1m+1[√−F(r,t)c7+c264c27−c62c7]−12(β+1)−1m+12β+12(β+1)+1m+1+c6(2β+12(β+1)+1m+1)12(β+1)−1m+1[√−F(R0,t)c7+c264c27−c62c7]−12(β+1)−1m+12β+12(β+1)+1m+1≤ln(R0r). | (4.21) |
Dropping the first and fourth terms on the left of (4.21), we obtain
√−F(r,t)c7+c264c27≤[c8ln(rR0)−Q2(R0,t)]−2β+12(β+1)+1m+112(β+1)−1m+1+c62c7, | (4.22) |
where Q2(R0,t)=2c7(12(β+1)−1m+1)c6(β2(β+1)+2m+1)[√−F(R0,t)c7+c264c27−c62c7]β2(β+1)+2m+12β+12(β+1)+1m+1 and c8=12(β+1)−1m+1c6(2β+12(β+1)+1m+1).
Squaring (4.22) we have
−F(r,t)≤c7[c8ln(rR0)−Q2(R0,t)]−2β+1β+1+2m+112(β+1)−1m+1+c6[c8ln(rR0)−Q2(R0,t)]−2β+12(β+1)+1m+112(β+1)−1m+1. | (4.23) |
From (4.15) and (4.23) we can obtain the following theorem.
Theorem 4.1. Let (u,v) be the solution of (1.1), (1.2), (2.1) and (2.2) with ρ(q2)=b1+b2q2β, where 12(β+1)>max{1m+1,1n+1}. Then for fixed t, when r→∞, (u,v) either grows algebraically or decays logarithmically. The growth rate is at least as fast as z1c5 and the decay rate is at least as fast as (lnr)−2β+12(β+1)+1m+112(β+1)−1m+1.
Remark 4.1. Obviously, in this case of ρ(q2)=b1+b2q2β, the decay rate obtained by Theorem 4.1 is slower than that obtained by Theorem 2.1 and Theorem 3.1.
In this section, we concern with a system of two coupled viscoelastic equations
utt−Δu+∫t0h1(t−η)Δu(η)dη+f1(u,v)=0, | (5.1) |
vtt−Δv+∫t0h2(t−η)Δv(η)dη+f2(u,v)=0, | (5.2) |
which describes the interaction between two different fields arising in viscoelasticity. In (5.1) and (5.2), 0<t<T and h1,h2 are differentiable functions satisfying h1(0),h2(0)>0 and
2(∫T0h21(T−η)dη), 4Th1(0)[∫T0(h′1(T−η)−ωh1(T−η))2dη]≤h1(0), | (5.3) |
2(∫T0h22(T−η)dη), 4Th2(0)[∫T0(h′2(T−η)−ωh2(T−η))2dη]≤h2(0). | (5.4) |
Messaoudi and Tatar [24] considered the system (5.1) and (5.2) in a bounded domain and proved the uniform decay for the solution when t→∞. For more special cases, one can refer to [25,26,27]. They mainly concerned the well-posedness of the solutions and proved that the solutions decayed uniformly under some suitable conditions. However, the present paper extends the previous results to Eqs (5.1) and (5.2) in an exterior region. We consider Eqs (5.1) and (5.2) with the initial-boundary conditions (2.1) and (2.2) in Ω.
We define two functions
G1(r,t)=∫t0∫B(r)e−ωη∇u⋅xruηdsdη−∫t0∫B(r)e−ωη(∫η0h1(η−s)∇uds)⋅xruηdsdη≐I1+I2, | (5.5) |
G2(r,t)=∫t0∫B(r)e−ωη∇v⋅xrvηdsdη−∫t0∫B(r)e−ωη(∫η0h2(η−s)∇vds)⋅xrvηdsdη≐J1+J2. | (5.6) |
Integrating (5.5) from r0 to r and using (5.1), (5.2), (2.1), (2.2) and the divergence theorem, we have
G1(r,t)=G1(r0,t)+12∫rr0∫B(ξ)e−ωt[|ut|2+|∇u|2]dsdξ+12ω∫t0∫rr0∫B(ξ)e−ωη|uη|2dsdξdη+12ω∫t0∫rr0∫B(ξ)e−ωη|∇u|2dsdξdη+∫t0∫rr0∫B(ξ)e−ωηh1(0)|∇u|2dsdξdη−∫rr0∫B(ξ)e−ωt(∫t0h1(t−τ)∇udτ)⋅∇udsdξ+∫t0∫rr0∫B(ξ)e−ωη[∫η0(h′1(η−τ)−ωh1(η−τ))∇udτ]⋅∇udsdξdη+∫t0∫rr0∫B(ξ)e−ωηf1(u,v)uηdsdξdη. | (5.7) |
From (5.7) it follows that
∂∂zG1(r,t)=12∫B(r)e−ωt[|ut|2+|∇u|2]ds+12ω∫t0∫B(r)e−ωη|uη|2dsdη+12ω∫t0∫B(r)e−ωη|∇u|2dsdη+∫t0∫B(r)e−ωηh1(0)|∇u|2dsdη−∫B(r)e−ωt(∫t0h1(t−τ)∇udτ)⋅∇uds+∫t0∫B(r)e−ωη[∫η0(h′1(η−τ)−ωh1(η−τ))∇udτ]⋅∇udsdη+∫t0∫B(r)e−ωηf1(u,v)uηdsdη. | (5.8) |
By the Young inequality and the Hölder inequality, we have
|−∫B(r)e−ωt(∫t0h1(η−τ)∇udτ)⋅∇uds|≤∫B(r)e−ωt[(∫t0h1(t−s)∇u(s)ds)2+14|∇u|2]ds≤(∫t0h21(t−τ)dτ)∫t0∫B(r)e−ωη|∇u|2dsdη+14∫B(r)e−ωt|∇u|2ds, | (5.9) |
and
|∫t0∫B(r)e−ωη[∫η0(h′1(η−τ)−ωh1(η−τ))∇udτ]⋅∇udsdη|≤th1(0)[∫t0(h′1(η−τ)−ωh1(η−τ))2dτ]∫t0∫B(r)e−ωη|∇u|2dsdη+14∫t0∫B(r)e−ωηh1(0)|∇u|2dsdη. | (5.10) |
Inserting (5.9) and (5.10) into (5.8) and using (5.3), we have
∂∂zG1(r,t)≥12∫B(r)e−ωt[|ut|2+12|∇u|2]ds+12ω∫t0∫B(r)e−ωη|uη|2dsdη+12ω∫t0∫B(r)e−ωη|∇u|2dsdη+∫t0∫B(r)e−ωηf1(u,v)uηdsdη. | (5.11) |
Similar to (5.11), we also have for G2(r,t)
∂∂zG2(r,t)≥12∫B(r)e−ωt[|vt|2+12|∇v|2]ds+12ω∫t0∫B(r)e−ωη|vη|2dsdη+12ω∫t0∫B(r)e−ωη|∇v|2dsdη+∫t0∫B(r)e−ωηf2(u,v)vηdsdη. | (5.12) |
If we define
G(r,t)=G1(r,t)+G2(r,t), |
then by (5.11) and (5.12) we have
∂∂zG(r,t)≥12∫B(r)e−ωt[|ut|2+|vt|2+12|∇u|2+12|∇v|2+2F(u,v)]ds+12ω∫t0∫B(r)e−ωη[|uη|2+|vη|2+|∇u|2+|∇v|2+2F(u,v)]dsdη. | (5.13) |
On the other hand, we bound G(r,t) by ∂∂rG(r,t). Using the Hölder inequality, the AG mean inequality, (5.3) and combining (5.13), we have
|I1|+|J1|≤(∫t0∫B(r)e−ωη|∇u|2dAdη⋅∫t0∫B(r)e−ωη|uη|2dsdη)12+(∫t0∫B(r)e−ωη|∇v|2dAdη⋅∫t0∫B(r)e−ωη|vη|2dsdη)12≤1ω[∂∂rG(r,t)], | (5.14) |
and
|I2|≤(∫t0∫B(r)e−ωη|∫η0h1(η−τ)∇udτ|2dsdη⋅∫t0∫B(r)e−ωη|uη|2dsdη)12≤(∫t0∫B(r)e−ωη(∫η0h21(η−τ)dτ)(∫η0|∇u|2dτ)dsdη⋅∫t0∫B(r)e−ωη|uη|2dsdη)12≤t(∫t0h21(η−τ)dτ)12(∫t0∫B(r)e−ωη|∇u|2dsdη⋅∫t0∫B(r)e−ωη|uη|2dsdη)12≤t2(∫t0h21(η−τ)dτ)12[∫t0∫B(r)e−ωη|∇u|2dsdη+∫t0∫B(r)e−ωη|uη|2dsdη]≤T2ω√h1(0)[∂∂rG(r,t)]. | (5.15) |
Similar to (5.15), we have
|J2|≤T2ω√h2(0)[∂∂rG(r,t)]. | (5.16) |
Inserting (5.14)–(5.16) into (5.5) and (5.6), we have
|G(r,t)|≤c91ω[∂∂rG(r,t)], | (5.17) |
where c9=T2(√h1(0)+√h2(0))+2.
We can follow the similar arguments given in the previous sections to obtain the following theorem.
Theorem 5.1. Let (u,v) be the solution of (5.1), (5.2), (2.1) and (2.2) in Ω, and (5.3) and (5.4) hold. For fixed t,
(1) If ∃ R0≥r0, G(R0,t)≥0, then
G(r,t)≥G(R0,t)eωb9(r−R0). |
(2) If ∀ r≥r0, G(r,t)<0, then
−G(r,t)≤[−G(r0,t)]e−ωb9(r−r0). |
Again, the rate of growth or decay obtained in this case is arbitrarily large
Remark 5.1. It is clear that the above analysis can be adapted without difficulties to the equation (see [28,29])
utt−k0△u+∫t0div[a(x)h(t−s)∇u(s)]ds+b(x)g(ut)+f(u)=0 |
and the equation (see [30])
|ut|σutt−k0△u−Δutt+∫t0h(t−s)Δu(s)]ds−γΔut=0 |
with some suitable g and a(x)+b(x)≥b10>0 and k0,σ,γ>0.
In this paper, we have considered several situations where the solutions of Eqs (1.1) and (1.2) either grow or decay exponentially or polynomially. We emphasize that the Poincaré inequality on the cross sections is not used in this paper. Thus, our results also hold for the two-dimensional case. On the other hand, there are some deeper problems to be studied in this paper. We can continue to study the continuous dependence of coefficients in the equation as that in [31]. These are the issues we will continue to study in the future.
The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.
The authors express their heartfelt thanks to the editors and referees who have provided some important suggestions. This work was supported by the Guangzhou Huashang College Tutorial Scientific Research Project (2022HSDS09), National Natural Science Foundation of China (11371175) and the Research Team Project of Guangzhou Huashang College (2021HSKT01).
The authors declare that there is no conflict of interest.
[1] |
Ahmadi Z, Martens P, Koch C, et al. (2018) Towards Bankruptcy Prediction: Deep Sentiment Mining to Detect Financial Distress from Business Management Reports. 2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA): 293–302. doi: 10.1109/DSAA.2018.00040
![]() |
[2] |
Aljawazneh H, Mora AM, García-Sánchez P, et al. (2021) Comparing the Performance of Deep Learning Methods to Predict Companies' Financial Failure. IEEE Access 9: 97010–97038. doi: 10.1109/ACCESS.2021.3093461
![]() |
[3] |
Antweiler W, Frank MZ (2004) Is All That Talk Just Noise? The Information Content of Internet Stock Message Boards. J Financ 59: 1259–1294. doi: 10.1111/j.1540-6261.2004.00662.x
![]() |
[4] |
Baek Y, Kim HY (2018) ModAugNet: A new forecasting framework for stock market index value with an overfitting prevention LSTM module and a prediction LSTM module. Expert Syst Appl 113: 457–480. doi: 10.1016/j.eswa.2018.07.019
![]() |
[5] |
Bahmani-Oskooee M, Saha S (2019) On the effects of policy uncertainty on stock prices: an asymmetric analysis. Quant Financ Econ 3: 412–424. doi: 10.3934/QFE.2019.2.412
![]() |
[6] |
Baranoff EG, Brockett P, Sager TW (2019) Was the U.S. life insurance industry in danger of systemic risk by using derivative hedging prior to the 2008 financial crisis? Quant Financ Econ 3: 145–164. doi: 10.3934/QFE.2019.1.145
![]() |
[7] |
Becerra-Vicario R, Alaminos D, Aranda E, et al. (2020) Deep Recurrent Convolutional Neural Network for Bankruptcy Prediction: A Case of the Restaurant Industry. Sustainability 12: 5180. doi: 10.3390/su12125180
![]() |
[8] |
Buxton E, Kriz K, Cremeens M, et al. (2019) An Auto Regressive Deep Learning Model for Sales Tax Forecasting from Multiple Short Time Series. 2019 18th IEEE International Conference on Machine Learning and Applications (ICMLA), 1359–1364. doi: 10.1109/ICMLA.2019.00221
![]() |
[9] | Cao Y, Shao Y, Zhang H (2021) Study on early warning of E-commerce enterprise financial risk based on deep learning algorithm. Electron Commer Res. |
[10] | Cao M, Qiu G, Lin X (2021) Research on big data financial risk behavior prediction based on deep learning. J Xinyang Agri Forestry Univ 31: 119–122. |
[11] |
Castelvecchi D (2016) Can we open the black box of AI? Nature News 538: 20–23. doi: 10.1038/538020a
![]() |
[12] | Chen G, Ju C, Peng X (2018) The Knn-smote-lstm based credit card fraud risk detection network model (in Chinese). Zhejiang: Zhejiang Gongshang University. |
[13] | Cohen S (2021) The basics of machine learning: strategies and techniques, In: Cohen S. Editor, Artificial Intelligence and Deep Learning in Pathology, Amsterdam: Elsevier, 13–40. |
[14] | Davis D (2019) AI Unleashes the Power of Unstructured Data, 2019. Available from: https://www.cio.com/article/3406806/ai-unleashes-the-power-of-unstructured-data.html. |
[15] | Devlin J, Chang MW, Lee K, et al. (2018) Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv: 1810.04805. |
[16] | Erica (2019) Introduction to the Fundamentals of Panel Data, 2019. Available from: http://www.aptech.com/blog/introduction-to-the-fundamentals-of-panel-data. |
[17] |
Feng W, Guan N, Li Y, et al. (2017) Audio visual speech recognition with multimodal recurrent neural networks. 2017 International Joint Conference on Neural Networks (IJCNN): 681–688. doi: 10.1109/IJCNN.2017.7965918
![]() |
[18] |
Fischer T, Krauss C (2018) Deep learning with long short-term memory networks for financial market predictions. Eur J Oper Res 270: 654–669. doi: 10.1016/j.ejor.2017.11.054
![]() |
[19] | González-Carvajal S, Garrido-Merchán EC (2020) Comparing BERT against traditional machine learning text classification. arXiv preprint arXiv: 1911.03829. |
[20] | Gopika P, Sowmya V, Gopalakrishnan EA, et al. (2020) Transferable approach for cardiac disease classification using deep learning, In: Agarwal B, Balas VE, Jain LC, et al. Editors, Deep Learning Techniques for Biomedical and Health Informatics, Cham: Springer, 285–303. |
[21] |
Havierniková K, Kordoš M (2019) The SMEs' perception of financial risks in the context of cluster cooperation. Quant Financ Econ 3: 586–607. doi: 10.3934/QFE.2019.3.586
![]() |
[22] | Haykin S (2000) Neural Networks: A Guided Tour, In: Sinha NK, Gupta MM. Editors, Soft Computing and Intelligent Systems: Theory and Applications, New York: Academic Press, 71–80. |
[23] |
He K, Ji L, Tso GKF, et al. (2018) Forecasting Exchange Rate Value at Risk using Deep Belief Network Ensemble based Approach. Procedia Comput Sci 139: 25–32. doi: 10.1016/j.procs.2018.10.213
![]() |
[24] | Hiew JZG, Huang X, Mou H, et al. (2019) BERT-based financial sentiment index and LSTM-based stock return predictability. arXiv preprint arXiv: 1906.09024. |
[25] |
Hosaka T (2019) Bankruptcy prediction using imaged financial ratios and convolutional neural networks. Expert Syst Appl 117: 287–299. doi: 10.1016/j.eswa.2018.09.039
![]() |
[26] |
Hsieh TJ, Hsiao HF, Yeh WC (2011) Forecasting stock markets using wavelet transforms and recurrent neural networks: An integrated system based on artificial bee colony algorithm. Appl Soft Comput 11: 2510–2525. doi: 10.1016/j.asoc.2010.09.007
![]() |
[27] |
Jan C (2021) Financial Information Asymmetry: Using Deep Learning Algorithms to Predict Financial Distress. Symmetry 13: 443. doi: 10.3390/sym13030443
![]() |
[28] |
Jan C (2021) Detection of Financial Statement Fraud Using Deep Learning for Sustainable Development of Capital Markets under Information Asymmetry. Sustainability 13: 9879. doi: 10.3390/su13179879
![]() |
[29] |
Jang Y, Jeong I, Cho YK (2020) Business Failure Prediction of Construction Contractors Using a LSTM RNN with Accounting, Construction Market, and Macroeconomic Variables. J Manage Eng 36: 04019039. doi: 10.1061/(ASCE)ME.1943-5479.0000733
![]() |
[30] |
Jiang M, Liu J, Zhang L, et al. (2020) An improved Stacking framework for stock index prediction by leveraging tree-based ensemble models and deep learning algorithms. Phys A 541: 122272. doi: 10.1016/j.physa.2019.122272
![]() |
[31] |
Jiang W (2021) Applications of deep learning in stock market prediction: recent progress. Expert Syst Appl 184: 115537. doi: 10.1016/j.eswa.2021.115537
![]() |
[32] |
Kim A, Yang Y, Lessmann S, et al. (2020) Can deep learning predict risky retail investors? A case study in financial risk behavior forecasting. Eur J Oper Res 283: 217–234. doi: 10.1016/j.ejor.2019.11.007
![]() |
[33] |
Kim SH, Kim D (2014) Investor sentiment from internet message postings and the predictability of stock returns. J Econ Behav Organ 107: 708–729. doi: 10.1016/j.jebo.2014.04.015
![]() |
[34] |
Lang M, Stice-Lawrence L (2015) Textual analysis and international financial reporting: Large sample evidence. J Account Econ 60: 110–135. doi: 10.1016/j.jacceco.2015.09.002
![]() |
[35] |
Lecun Y, Bottou L, Bengio Y, et al. (1998) Gradient-based learning applied to document recognition. Proceed IEEE 86: 2278–2324. doi: 10.1109/5.726791
![]() |
[36] |
Leo M, Sharma S, Maddulety K (2019) Machine learning in banking risk management: A literature review. Risks 7: 29. doi: 10.3390/risks7010029
![]() |
[37] |
Li S, Shi W, Wang J, et al. (2021) A Deep Learning-Based Approach to Constructing a Domain Sentiment Lexicon: a Case Study in Financial Distress Prediction. Inf Process Manage 58: 102673. doi: 10.1016/j.ipm.2021.102673
![]() |
[38] |
Linsmeier TJ, Pearson ND (2000) Value at Risk. Financ Anal J 56: 47–67. doi: 10.2469/faj.v56.n2.2343
![]() |
[39] |
Liu K, Luo C, Li Z (2019) Investigating the risk spillover from crude oil market to BRICS stock markets based on Copula-POT-CoVaR models. Quant Financ Econ 3: 754–771. doi: 10.3934/QFE.2019.4.754
![]() |
[40] | Lv H (2019) International experience and enlightenment of financial foundation database construction. Financ Perspect J, 67–74. |
[41] |
Ma X, Lv S (2019) Financial credit risk prediction in internet finance driven by machine learning. Neural Comput Appl 31: 8359–8367. doi: 10.1007/s00521-018-3963-6
![]() |
[42] |
Mai F, Tian S, Lee C, et al. (2019) Deep learning models for bankruptcy prediction using textual disclosures. Eur J Oper Res 274: 743–758. doi: 10.1016/j.ejor.2018.10.024
![]() |
[43] |
Nicola M, Alsafi Z, Sohrabi C, et al. (2020) The socio-economic implications of the coronavirus pandemic (COVID-19): A review. Int J Surg 78: 185–193. doi: 10.1016/j.ijsu.2020.04.018
![]() |
[44] | Nisha SS, Sathik MM, Meeral MN (2021) Application, algorithm, tools directly related to deep learning, In: Balas VE, Mishra BK, Kumar R. Editors, Handbook of Deep Learning in Biomedical Engineering, Amsterdam: Elsevier, 61–84. |
[45] |
Ozbayoglu AM, Gudelek MU, Sezer OB (2020) Deep learning for financial applications: A survey. Appl Soft Comput 93: 106384. doi: 10.1016/j.asoc.2020.106384
![]() |
[46] |
Peng Y, Wang G, Kou G, et al. (2011) An empirical study of classification algorithm evaluation for financial risk prediction. Appl Soft Comput 11: 2906–2915. doi: 10.1016/j.asoc.2010.11.028
![]() |
[47] |
Qu Y, Quan P, Lei M, et al. (2019) Review of bankruptcy prediction using machine learning and deep learning techniques. Procedia Comput Sci 162: 895–899. doi: 10.1016/j.procs.2019.12.065
![]() |
[48] | Reimers C, Requena-Mesa C (2020) Deep Learning—an Opportunity and a Challenge for Geo- and Astrophysics, In: Škoda P, Adam F. Editors, Knowledge Discovery in Big Data from Astronomy and Earth Observation, Amsterdam: Elsevier, 251–265. |
[49] | Roshan S, Srivathsan G, Deepak K, et al. (2020) Violence Detection in Automated Video Surveillance: Recent Trends and Comparative Studies, In: Peter D, Alavi AH, Javadi B, et al. Editors, The Cognitive Approach in Cloud Computing and Internet of Things Technologies for Surveillance Tracking Systems, Amsterdam: Elsevier, 157–171. |
[50] |
Sezer OB, Gudelek MU, Ozbayoglu AM (2020) Financial time series forecasting with deep learning: A systematic literature review: 2005–2019. Appl Soft Comput 90: 106181. doi: 10.1016/j.asoc.2020.106181
![]() |
[51] | Sirignano J, Sadhwani A, Giesecke K (2016) Deep learning for mortgage risk. arXiv preprint arXiv: 1607.02470. |
[52] | Skryjomski P, Krawczyk B (2017) Influence of minority class instance types on SMOTE imbalanced data oversampling. Proc Machine Learn Res 74: 7–21. |
[53] |
Smiti S, Soui M (2020) Bankruptcy Prediction Using Deep Learning Approach Based on Borderline SMOTE. Infor Syst Front 22: 1067–1083. doi: 10.1007/s10796-020-10031-6
![]() |
[54] | Song Y, Cai W (2021) Visual feature representation in microscopy image classification, In: Chen M. Editor, Computer Vision for Microscopy Image Analysis, Amsterdam: Elsevier, 73–100. |
[55] |
Song Y, Peng Y (2019) A MCDM-Based Evaluation Approach for Imbalanced Classification Methods in Financial Risk Prediction. IEEE Access 7: 84897–84906. doi: 10.1109/ACCESS.2019.2924923
![]() |
[56] | Sousa MG, Sakiyama K, Rodrigues LS, et al. (2019) BERT for Stock Market Sentiment Analysis. 2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI): 1597–1601. |
[57] |
Sukharev OS (2020) Structural analysis of income and risk dynamics in models of economic growth. Quant Financ Econ 4: 1–18. doi: 10.3934/QFE.2020001
![]() |
[58] |
Tserng HP, Chen PC, Huang WH, et al. (2014) Prediction of default probability for construction firms using the logit model. J Civil Eng Manage 20: 247–255. doi: 10.3846/13923730.2013.801886
![]() |
[59] |
Tsukioka Y, Yanagi J, Takada T (2018) Investor sentiment extracted from internet stock message boards and IPO puzzles. Int Rev Econ Financ 56: 205–217. doi: 10.1016/j.iref.2017.10.025
![]() |
[60] | Wang J, Xin F, Zhang W (2019) Research on bank green credit risk evaluation based on BP Neural Network (in Chinese). Jinan: University of Jinan. |
[61] |
Wang KJ, Adrian AM, Chen KH, et al. (2015). A hybrid classifier combining Borderline-SMOTE with AIRS algorithm for estimating brain metastasis from lung cancer: A case study in Taiwan. Comput Methods Programs Biomed 119: 63–76. doi: 10.1016/j.cmpb.2015.03.003
![]() |
[62] | Wang Q, Yi B, Xie Y, et al. (2003) The hardware structure design of perceptron with FPGA implementation. 2003 IEEE International Conference on Systems, Man and Cybernetics, 762–767. |
[63] |
Wang W, Liu M, Zhang Y, et al. (2019) Financial numeral classification model based on BERT. NTCIR 2019: NII Testbeds and Community for Information Access Research, 193–204. doi: 10.1007/978-3-030-36805-0_15
![]() |
[64] | Woo S, Lee CL (2018) Decision boundary formation of deep convolution networks with ReLU. 2018 IEEE 16th Intl Conf on Dependable, Autonomic and Secure Computing, 885–888. |
[65] | Wyrobek J (2018) Predicting Bankruptcy at Polish Companies: A Comparison of Selected Machine Learning and Deep Learning Algorithms. Zeszyty Naukowe Uniwersytetu Ekonomicznego w Krakowie, 41–60. |
[66] | Xiao J, Wang L, Li N, et al. (2021) Research on perceptual cognitive technology and its application in financial risk warning. CAAI Trans Intell Syst 16: 941–961. |
[67] |
Xing Y, Lv C, Wang H, et al. (2019) Driver Lane Change Intention Inference for Intelligent Vehicles: Framework, Survey, and Challenges. IEEE Trans Vehicular Technol 68: 4377–4390. doi: 10.1109/TVT.2019.2903299
![]() |
[68] |
Yang X, Mao S, Gao H, et al. (2019) Novel Financial Capital Flow Forecast Framework Using Time Series Theory and Deep Learning: A Case Study Analysis of Yu'e Bao Transaction Data. IEEE Access 7: 70662–70672. doi: 10.1109/ACCESS.2019.2919189
![]() |
[69] | Yu SL, Li Z (2018) Forecasting Stock Price Index Volatility with LSTM Deep Neural Network, In: Tavana M, Patnaik S. Editors, Recent Developments in Data Science and Business Analytics, Cham: Springer Nature, 265–272. |
[70] | Yu X (2017) Machine learning application in online lending risk prediction. arXiv preprint arXiv: 1707.04831. |
[71] | Zhang H (2020) A forecasting method of short-term stock volatility based on deep learning. Modern Mark (Business Edition) 05: 179–181. |
[72] |
Zhao D, Ding J, Chai S (2018) Systemic financial risk prediction using least squares support vector machines. Modern Phys Lett B 32: 1850183. doi: 10.1142/S021798491850183X
![]() |
[73] | Zhao J (2020) Corporate Financial Risk Prediction Based on Embedded System and Deep Learning. Microprocess Microsy, 103405. |
[74] |
Zhou L (2013) Performance of corporate bankruptcy prediction models on imbalanced dataset: The effect of sampling methods. Knowledge-Based Syst 41: 16–25. doi: 10.1016/j.knosys.2012.12.007
![]() |
[75] | Zhou X (2017) Holding the bottom line of avoiding systemic financial risks, 2017. Available From: http://cpc.people.com.cn/n1/2017/1122/c64094-29660265.html. |
1. | S.A. Pedro, J.M. Tchuenche, HIV/AIDS dynamics: Impact of economic classes with transmission from poor clinical settings, 2010, 267, 00225193, 471, 10.1016/j.jtbi.2010.09.019 | |
2. | C. P. Bhunu, S. Mushayabasa, H. Kojouharov, J. M. Tchuenche, Mathematical Analysis of an HIV/AIDS Model: Impact of Educational Programs and Abstinence in Sub-Saharan Africa, 2011, 10, 1570-1166, 31, 10.1007/s10852-010-9134-0 | |
3. | Sarada S. Panchanathan, Diana B. Petitti, Douglas B. Fridsma, The development and validation of a simulation tool for health policy decision making, 2010, 43, 15320464, 602, 10.1016/j.jbi.2010.03.013 | |
4. | Cristiana J. Silva, Delfim F.M. Torres, A SICA compartmental model in epidemiology with application to HIV/AIDS in Cape Verde, 2017, 30, 1476945X, 70, 10.1016/j.ecocom.2016.12.001 | |
5. | S. D. Hove-Musekwa, F. Nyabadza, H. Mambili-Mamboundou, Modelling Hospitalization, Home-Based Care, and Individual Withdrawal for People Living with HIV/AIDS in High Prevalence Settings, 2011, 73, 0092-8240, 2888, 10.1007/s11538-011-9651-7 | |
6. | Emmanuelina L. Kateme, Jean M. Tchuenche, Senelani D. Hove-Musekwa, HIV/AIDS Dynamics with Three Control Strategies: The Role of Incidence Function, 2012, 2012, 2090-5572, 1, 10.5402/2012/864795 | |
7. | F. Nyabadza, Z. Mukandavire, S.D. Hove-Musekwa, Modelling the HIV/AIDS epidemic trends in South Africa: Insights from a simple mathematical model, 2011, 12, 14681218, 2091, 10.1016/j.nonrwa.2010.12.024 | |
8. | Xiaodong Wang, Chunxia Wang, Kai Wang, Extinction and persistence of a stochastic SICA epidemic model with standard incidence rate for HIV transmission, 2021, 2021, 1687-1847, 10.1186/s13662-021-03392-y |