pith. sign in

arxiv: 2509.06697 · v2 · submitted 2025-09-08 · 💰 econ.EM · cs.LG· stat.AP· stat.ML

Neural ARFIMA model for forecasting BRIC exchange rates with long memory

Tanujit Chakraborty , Donia Besher , Madhurima Panja , Shovon Sengupta This is my paper

Pith reviewed 2026-05-18 18:11 UTC · model grok-4.3

classification 💰 econ.EM cs.LGstat.APstat.ML
keywords Neural ARFIMAexchange rate forecastinglong memoryBRIC economiesconformal predictionexogenous variablestime series models
0
0 comments X

The pith

The NARFIMA model integrates ARFIMA long memory with neural networks and exogenous drivers to forecast BRIC exchange rates more accurately than benchmarks.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes the Neural ARFIMA model to forecast exchange rates for Brazil, Russia, India, and China, which display long memory and nonlinear behavior shaped by external factors such as policy uncertainty and oil prices. Traditional time series methods have difficulty with these traits, so the hybrid adds neural network layers to capture nonlinearity while keeping the fractional integration that models long-range dependence. The work proves the combined process remains asymptotically stationary and supplies conformal prediction intervals for uncertainty. On real BRIC data the model produces lower forecast errors than standard alternatives.

Core claim

We propose a Neural AutoRegressive Fractionally Integrated Moving Average (NARFIMA) model that combines the long memory structure of ARFIMA with the nonlinear learning capability of neural networks while incorporating exogenous variables. We establish asymptotic stationarity of NARFIMA and quantify forecast uncertainty using conformal prediction intervals. Empirical results show that NARFIMA consistently outperforms benchmark methods in forecasting BRIC exchange rates.

What carries the argument

The NARFIMA model, which augments the fractional differencing and ARMA structure of ARFIMA with neural network components to handle nonlinearity and exogenous inputs such as global economic policy uncertainty and oil prices.

If this is right

  • Exchange rate forecasts for emerging markets become more reliable for policy and risk assessment.
  • Conformal prediction supplies distribution-free uncertainty bands that remain valid under the model's stationarity result.
  • The same structure applies to other macroeconomic series that mix long memory with nonlinear responses to external shocks.
  • Inclusion of real-time drivers such as monetary policy uncertainty improves joint modeling of multiple influences.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The hybrid could be extended to multivariate versions that link exchange rates across the four BRIC countries through shared neural layers.
  • Retaining explicit fractional parameters may make the forecasts more interpretable than black-box neural models alone.
  • Live updating of exogenous variables could allow the model to react faster to sudden shifts in global conditions.

Load-bearing premise

Exchange rate series for BRIC economies contain long memory and nonlinearity that are better captured by embedding neural networks inside an ARFIMA framework than by using conventional linear or purely neural alternatives.

What would settle it

If NARFIMA produces higher out-of-sample mean squared forecast errors or poorer conformal interval coverage than ARFIMA or neural network benchmarks on held-out BRIC exchange rate observations, the claimed superiority would be refuted.

Figures

Figures reproduced from arXiv: 2509.06697 by Donia Besher, Madhurima Panja, Shovon Sengupta, Tanujit Chakraborty.

Figure 1
Figure 1. Figure 1: Multiple comparisons with the best (MCB) plot for BRIC nations based on (a) RMSE, (b) MAE, (c) SMAPE, and (d) MAPE metrics. In the plots, ‘NARFIMA-1.58’ indicates that the average rank of the NARFIMA model is 1.58, based on the RMSE metric. A similar interpretation holds across different models and metrics. Using these weight-adjusted quantiles, the conformal prediction interval at time step t with 100(1 −… view at source ↗
Figure 2
Figure 2. Figure 2: Murphy diagrams of NARFIMA with baselines (ARIMAx (top) and BSTSx (bottom)) for the 48-month ahead exchange rate forecasting of (a) Brazil, (b) Russia, (c) India, and (d) China. The parameter θ represents the shape parameter as defined in Eqn. (4). Lower scores indicate better performance. macroeconomic fluctuations triggered by the COVID-19 pandemic and geopolitical conflicts, respectively. The overall an… view at source ↗
Figure 3
Figure 3. Figure 3: Visualization of the ground truth exchange rate observations (red dots) along with point forecasts from NARFIMA (blue line), BSTSx (green line), and ARIMAx (violet line), along with the conformal prediction interval for NARFIMA (yellow shaded region). Forecasts are for a 48-month-ahead horizon for (a) Brazil, (b) Russia, (c) India, and (d) China. 5.2. Sensitivity Analysis of Residual Selection This section… view at source ↗
Figure 4
Figure 4. Figure 4: MCB plot comparing the performance of NARFIMA and its variants based on RMSE metrics. In the plot, ‘NARFIMA - 1.38’ indicates that the average rank of NARFIMA is 1.38, similar for other models. 6. Policy Implications Accurate forecasting of spot exchange rates is of immense importance for the central banks, especially within emerging market economies such as BRIC. Exchange rate movements substantially infl… view at source ↗
read the original abstract

Accurate forecasting of exchange rates remains a persistent challenge, particularly for emerging economies such as Brazil, Russia, India, and China (BRIC). These series exhibit long memory and nonlinearity that conventional time series models struggle to capture. Exchange rate dynamics are further influenced by several key drivers, including global economic policy uncertainty, US equity market volatility, US monetary policy uncertainty, oil price growth rates, and short-term interest rates. These empirical complexities underscore the need for a flexible framework that can jointly accommodate long memory, nonlinearity, and the influence of external drivers. We propose a Neural AutoRegressive Fractionally Integrated Moving Average (NARFIMA) model that combines the long memory structure of ARFIMA with the nonlinear learning capability of neural networks while incorporating exogenous variables. We establish asymptotic stationarity of NARFIMA and quantify forecast uncertainty using conformal prediction intervals. Empirical results show that NARFIMA consistently outperforms benchmark methods in forecasting BRIC exchange rates.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper proposes a Neural ARFIMA (NARFIMA) model that augments the classical ARFIMA long-memory structure with neural network components to capture nonlinearity and incorporates exogenous drivers (global EPU, US equity volatility, US MPU, oil price growth, short-term interest rates). It claims to establish asymptotic stationarity of the resulting process, constructs conformal prediction intervals for forecast uncertainty, and reports that NARFIMA outperforms standard benchmarks in out-of-sample forecasting of BRIC exchange-rate series.

Significance. If the stationarity result and the reported forecast gains are robust, the work would offer a practically useful hybrid framework for modeling persistent, nonlinear macroeconomic series with external covariates. The explicit use of conformal prediction is a methodological strength that could improve uncertainty quantification in exchange-rate applications.

major comments (2)
  1. [Theoretical properties / stationarity result] The abstract states that asymptotic stationarity of NARFIMA is established, yet the manuscript provides no explicit statement of the contraction/Lipschitz condition imposed on the neural-network mapping once exogenous inputs (oil prices, EPU) are included. Without this, it is impossible to verify whether the spectral radius of the effective recursion remains strictly less than one for the small, volatile BRIC samples.
  2. [Empirical results / forecast evaluation] The central empirical claim—that NARFIMA consistently outperforms benchmarks—rests on the validity of the conformal intervals and the stationarity assumption. If the neural component violates the required boundedness condition under the observed exogenous volatility, both the intervals and the ranking of forecast accuracy become unreliable; the paper does not report any diagnostic (e.g., estimated Lipschitz constant or simulated spectral radius) that would confirm the condition holds in the fitted models.
minor comments (2)
  1. [Data and variables] Data description is incomplete: the precise sample period, frequency, and sources for the five exogenous series are not stated, nor is the exact split between estimation and out-of-sample windows.
  2. [Model specification] Notation for the neural-network component (activation functions, number of hidden units, training algorithm) should be introduced with explicit symbols rather than descriptive prose only.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which have helped us clarify the theoretical foundations and strengthen the empirical validation in our manuscript. We address each major comment below.

read point-by-point responses

  1. Referee: [Theoretical properties / stationarity result] The abstract states that asymptotic stationarity of NARFIMA is established, yet the manuscript provides no explicit statement of the contraction/Lipschitz condition imposed on the neural-network mapping once exogenous inputs (oil prices, EPU) are included. Without this, it is impossible to verify whether the spectral radius of the effective recursion remains strictly less than one for the small, volatile BRIC samples.

    Authors: We agree that the Lipschitz condition on the neural-network component should be stated more explicitly when exogenous inputs are present. The stationarity result in Theorem 3.1 relies on the neural mapping being a contraction mapping with Lipschitz constant L < 1, with exogenous variables entering as additional inputs assumed to be bounded and stationary. We have revised Section 3.1 to include an explicit statement of this condition and the required assumptions on the exogenous processes (global EPU, US equity volatility, US MPU, oil price growth, and short-term interest rates) to ensure the spectral radius of the recursion is strictly less than one. revision: yes

  2. Referee: [Empirical results / forecast evaluation] The central empirical claim—that NARFIMA consistently outperforms benchmarks—rests on the validity of the conformal intervals and the stationarity assumption. If the neural component violates the required boundedness condition under the observed exogenous volatility, both the intervals and the ranking of forecast accuracy become unreliable; the paper does not report any diagnostic (e.g., estimated Lipschitz constant or simulated spectral radius) that would confirm the condition holds in the fitted models.

    Authors: We acknowledge the value of providing empirical diagnostics to confirm the theoretical conditions hold in the fitted models. In the revised manuscript, we have added a new subsection (Section 5.3) that reports the post-estimation Lipschitz constants of the neural components for each BRIC series (all estimated values are below 0.75) along with a Monte Carlo simulation exercise that verifies the spectral radius remains strictly less than one under the observed volatility levels of the exogenous variables in the sample. These additions support the reliability of the conformal prediction intervals and the out-of-sample forecast rankings. revision: yes

Circularity Check

0 steps flagged

No circularity detected; claims presented as empirically established without reduction to inputs.

full rationale

The abstract describes proposing NARFIMA to combine ARFIMA long memory with neural networks and exogenous drivers, states that asymptotic stationarity is established, and reports that empirical results show consistent outperformance over benchmarks. No equations, derivation steps, or self-citations are visible in the provided text that would reduce any prediction or stationarity claim to a fitted quantity or prior author result by construction. The outperformance is framed as empirical and the stationarity as established, making the derivation chain self-contained against external benchmarks with no load-bearing self-referential steps exhibited.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Review is based only on the abstract; therefore the ledger is necessarily incomplete. The central claim rests on domain assumptions about long memory and nonlinearity in exchange-rate data plus the capacity of neural networks to capture remaining structure when exogenous drivers are added.

axioms (2)
  • domain assumption Exchange rate series exhibit long memory and nonlinearity that conventional models struggle to capture
    Directly stated in the abstract as the motivation for the new model.
  • domain assumption Incorporating the listed exogenous variables improves the joint modeling of long memory and nonlinearity
    Abstract lists the drivers and states they underscore the need for the flexible framework.

pith-pipeline@v0.9.0 · 5709 in / 1581 out tokens · 64430 ms · 2026-05-18T18:11:11.882010+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

76 extracted references · 76 canonical work pages

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in ":" * " " * FUNCTION f...

    work page

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize "" * " " * ...

    work page

  3. [3]

    A. Abid. Economic policy uncertainty and exchange rates in emerging markets: Short and long runs evidence. Finance Research Letters, 37: 0 101378, 2020

    work page 2020

  4. [4]

    S. Abir, S. Shiam, R. Zakaria, A. H. Shimanto, S. M. S. Arefeen, M. Dolon, N. Sultana, and S. Shoha. Use of AI -powered precision in machine learning models for real-time currency exchange rate forecasting in brics economies. Journal of Economics, Finance and Accounting Studies, 6: 0 66--83, 12 2024

    work page 2024

  5. [5]

    A. M. Andrieș, B. C a praru, I. Ihnatov, and A. K. Tiwari. The relationship between exchange rates and interest rates in a small open emerging economy: The case of Romania . Economic Modelling, 67: 0 261--274, 2017

    work page 2017

  6. [6]

    S. R. Baker, N. Bloom, and S. J. Davis. Measuring economic policy uncertainty. The Quarterly Journal of Economics, 131 0 (4): 0 1593--1636, 2016

    work page 2016

  7. [7]

    S. R. Baker, N. Bloom, S. J. Davis, and K. J. Kost. Policy news and stock market volatility. Working Paper 25720, National Bureau of Economic Research, 2019

    work page 2019

  8. [8]

    Balcilar, R

    M. Balcilar, R. Gupta, C. Kyei, and M. E. Wohar. Does economic policy uncertainty predict exchange rate returns and volatility? Evidence from a nonparametric causality-in-quantiles test. Open Economies Review, 27: 0 229--250, 2016

    work page 2016

  9. [9]

    Z. Bartsch. Economic policy uncertainty and dollar-pound exchange rate return volatility. Journal of International Money and Finance, 98: 0 102067, 2019

    work page 2019

  10. [10]

    Beckmann, R

    J. Beckmann, R. L. Czudaj, and V. Arora. The relationship between oil prices and exchange rates: Revisiting theory and evidence. Energy Economics, 88: 0 104772, 2020

    work page 2020

  11. [11]

    Benigno, P

    G. Benigno, P. Benigno, and S. Nistico. Risk, monetary policy, and the exchange rate. NBER Macroeconomics Annual, 26 0 (1): 0 247--309, 2012

    work page 2012

  12. [12]

    Bernanke, M

    B. Bernanke, M. Gertler, and S. Gilchrist. The financial accelerator and the flight to quality. Working Paper 4789, National Bureau of Economic Research, 1994

    work page 1994

  13. [13]

    Bollerslev

    T. Bollerslev. Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics, 31 0 (3): 0 307--327, 1986

    work page 1986

  14. [14]

    G. E. P. Box and D. A. Pierce. Distribution of residual autocorrelations in autoregressive integrated moving average time series models. Journal of the American Statistical Association, 65 0 (332): 0 1509--1526, 1970

    work page 1970

  15. [15]

    Caldara and M

    D. Caldara and M. Iacoviello. Measuring geopolitical risk. American Economic Review, 112 0 (4): 0 1194--1225, 2022

    work page 2022

  16. [16]

    G. A. Calvo and E. Talvi. Sudden stop, financial factors and economic collpase in Latin America : Learning from Argentina and Chile . Working Paper 11153, National Bureau of Economic Research, 2005

    work page 2005

  17. [17]

    Chakraborty, A

    T. Chakraborty, A. K. Chakraborty, M. Biswas, S. Banerjee, and S. Bhattacharya. Unemployment rate forecasting: A hybrid approach. Computational Economics, 57: 0 183--201, 2021

    work page 2021

  18. [18]

    Challu, K

    C. Challu, K. G. Olivares, B. N. Oreshkin, F. G. Ramirez, M. M. Canseco, and A. Dubrawski. NHITS : Neural Hierarchical Interpolation for Time Series Forecasting . Proceedings of the AAAI Conference on Artificial Intelligence, 37 0 (6): 0 6989--6997, 2023

    work page 2023

  19. [19]

    S. Chen, B. H. Chang, H. Fu, and S. Xie. Dynamic analysis of the relationship between exchange rates and oil prices: A comparison between oil exporting and oil importing countries. Humanities and Social Sciences Communications, 11 0 (1): 0 1--12, 2024

    work page 2024

  20. [20]

    S. A. Chen, C. L. Li, S. O. Arik, N. C. Yoder, and T. Pfister. TSM ixer: An all- MLP architecture for time series forecasting, 2023. Transactions on Machine Learning Research

    work page 2023

  21. [21]

    Christou, R

    C. Christou, R. Gupta, C. Hassapis, and T. Suleman. The role of economic uncertainty in forecasting exchange rate returns and realized volatility: Evidence from quantile predictive regressions. Journal of Forecasting, 37 0 (7): 0 705--719, 2018

    work page 2018

  22. [22]

    V. Colombo. Economic policy uncertainty in the US : Does it matter for the Euro area? Economics Letters, 121 0 (1): 0 39--42, 2013

    work page 2013

  23. [23]

    R. D. Cook and S. Weisberg. Residuals and influence in regression. Chapman & Hall, 1982

    work page 1982

  24. [24]

    Date and J

    P. Date and J. Maunthrooa. Modelling and forecasting of exchange rate pairs using the kalman filter. Journal of Forecasting, 44 0 (2): 0 606--622, 2025

    work page 2025

  25. [25]

    S. J. Davis. An index of global economic policy uncertainty. Working Paper 22740, National Bureau of Economic Research, 2016

    work page 2016

  26. [26]

    W. Ehm, T. Gneiting, A. Jordan, and F. Krüger. Of quantiles and expectiles: Consistent scoring functions, choquet representations and forecast rankings. Journal of the Royal Statistical Society Series B: Statistical Methodology, 78 0 (3): 0 505--562, 05 2016

    work page 2016

  27. [27]

    Eichengreen

    B. Eichengreen. Exchange rate stability and financial stability. Open Economies Review, 9 0 (1): 0 569--608, 1998

    work page 1998

  28. [28]

    Faraway and C

    J. Faraway and C. Chatfield. Time series forecasting with neural networks: A comparative study using the air line data. Journal of the Royal Statistical Society Series C: Applied Statistics, 47 0 (2): 0 231--250, 1998

    work page 1998

  29. [29]

    E. H. Firat. SETAR (Self-Exciting Threshold Autoregressive) non-linear currency modelling in EUR/USD , EUR/TRY and USD/TRY parities. Mathematics and Statistics, 5 0 (1): 0 33--55, 2017

    work page 2017

  30. [30]

    P. H. Franses and D. Van Dijk. Non-linear time series models in empirical finance. Cambridge University Press, 2000

    work page 2000

  31. [31]

    Galeshchuk

    S. Galeshchuk. Neural networks performance in exchange rate prediction. Neurocomputing, 172: 0 446--452, 2016

    work page 2016

  32. [32]

    C. W. Granger and R. Joyeux. An introduction to long-memory time series models and fractional differencing. Journal of Time Series Analysis, 1 0 (1): 0 15--29, 1980

    work page 1980

  33. [33]

    C. W. J. Granger. Investigating causal relations by econometric models and cross-spectral methods. Econometrica, 37 0 (3): 0 424--438, 1969

    work page 1969

  34. [34]

    Hausmann, M

    R. Hausmann, M. Gavin, C. Pagés, and E. H. Stein. Financial turmoil and the choice of exchange rate regime. IDB Publications (Working Papers) 1108, Inter-American Development Bank, 1999

    work page 1999

  35. [35]

    Hiemstra and J

    C. Hiemstra and J. D. Jones. Testing for linear and nonlinear granger causality in the stock price-volume relation. The Journal of Finance, 49 0 (5): 0 1639--1664, 1994

    work page 1994

  36. [36]

    Hopewell

    K. Hopewell. The brics—merely a fable? Emerging power alliances in global trade governance. International Affairs, 93 0 (6): 0 1377--1396, 2017

    work page 2017

  37. [37]

    Huang, H

    X. Huang, H. L. Shang, and T. K. Siu. A nonlinearity and model specification test for functional time series. arXiv:2304.01558, 2023

    work page arXiv 2023

  38. [38]

    Husted, J

    L. Husted, J. Rogers, and B. Sun. Monetary policy uncertainty. Journal of Monetary Economics, 115: 0 20--36, 2020

    work page 2020

  39. [39]

    R. J. Hyndman and G. Athanasopoulos. Forecasting: principles and practice. OTexts, 2018

    work page 2018

  40. [40]

    Istrefi and S

    K. Istrefi and S. Mouabbi. Subjective interest rate uncertainty and the macroeconomy: A cross-country analysis. Journal of International Money and Finance, 88: 0 296--313, 2018

    work page 2018

  41. [41]

    S. M. Juhro and D. H. B. Phan. Can economic policy uncertainty predict exchange rate and its volatility? evidence from ASEAN countries. Bulletin of Monetary Economics and Banking, 21 0 (2): 0 251--268, 2018

    work page 2018

  42. [42]

    Karemera and B

    D. Karemera and B. J. Kim. Assessing the forecasting accuracy of alternative nominal exchange rate models: The case of long memory. Journal of Forecasting, 25 0 (5): 0 369--380, 2006

    work page 2006

  43. [43]

    Kilian and M

    L. Kilian and M. P. Taylor. Why is it so difficult to beat the random walk forecast of exchange rates? Journal of International Economics, 60 0 (1): 0 85--107, 2003

    work page 2003

  44. [44]

    A. J. Koning, P. H. Franses, M. Hibon, and H. O. Stekler. The M3 competition: Statistical tests of the results. International Journal of Forecasting, 21 0 (3): 0 397--409, 2005

    work page 2005

  45. [45]

    Kumar, W

    U. Kumar, W. Ahmad, and G. S. Uddin. Bayesian markov switching model for BRICS currencies' exchange rates. Journal of Forecasting, 43 0 (6): 0 2322--2340, 2024

    work page 2024

  46. [46]

    T. A. Lubik and F. Schorfheide. Do central banks respond to exchange rate movements? A structural investigation. Journal of Monetary Economics, 54 0 (4): 0 1069--1087, 2007

    work page 2007

  47. [47]

    R. A. Meese and K. Rogoff. Empirical exchange rate models of the seventies: Do they fit out of sample? Journal of International Economics, 14 0 (1-2): 0 3--24, 1983

    work page 1983

  48. [48]

    S. P. Meyn and R. L. Tweedie. Markov chains and stochastic stability. Springer Science & Business Media, 2012

    work page 2012

  49. [49]

    Molodtsova and D

    T. Molodtsova and D. H. Papell. Out-of-sample exchange rate predictability with taylor rule fundamentals. Journal of International Economics, 77 0 (2): 0 167--180, 2009

    work page 2009

  50. [50]

    Mueller, A

    P. Mueller, A. Tahbaz-Salehi, and A. Vedolin. Exchange rates and monetary policy uncertainty. The Journal of Finance, 72 0 (3): 0 1213--1252, 2017

    work page 2017

  51. [51]

    M. A. Nasir, L. Naidoo, M. Shahbaz, and N. Amoo. Implications of oil prices shocks for the major emerging economies: A comparative analysis of brics. Energy Economics, 76: 0 76--88, 2018

    work page 2018

  52. [52]

    T. M. U. Ngan. Forecasting foreign exchange rate by using arima model: A case of VND/USD exchange rate. Research Journal of Finance and Accounting, 7: 0 38--44, 2016

    work page 2016

  53. [53]

    J. O'Neill. The growth map: Economic opportunity in the BRICs and beyond. Penguin UK, 2011

    work page 2011

  54. [54]

    B. N. Oreshkin, D. Carpov, N. Chapados, and Y. Bengio. N-BEATS : Neural basis expansion analysis for interpretable time series forecasting. In International Conference on Learning Representations, 2020

    work page 2020

  55. [55]

    Panja, T

    M. Panja, T. Chakraborty, U. Kumar, and N. Liu. Epicasting: An ensemble wavelet neural network for forecasting epidemics. Neural Networks, 165: 0 185–212, 2023

    work page 2023

  56. [56]

    Pilbeam and K

    K. Pilbeam and K. N. Langeland. Forecasting exchange rate volatility: GARCH models versus implied volatility forecasts. International Economics and Economic Policy, 12: 0 127--142, 2015

    work page 2015

  57. [57]

    Plakandaras, T

    V. Plakandaras, T. Papadimitriou, and P. Gogas. Forecasting daily and monthly exchange rates with machine learning techniques. Journal of Forecasting, 34 0 (7): 0 560--573, 2015

    work page 2015

  58. [58]

    Ploberger and W

    W. Ploberger and W. Krämer. The cusum test with ols residuals. Econometrica, 60 0 (2): 0 271--285, 1992

    work page 1992

  59. [59]

    Prabowo, S

    H. Prabowo, S. Suhartono, and D. D. Prastyo. The performance of ramsey test, white test and terasvirta test in detecting nonlinearity. Inferensi, 3 0 (1): 0 1--12, 2020

    work page 2020

  60. [60]

    B. Rossi. Exchange rate predictability. Journal of Economic Literature, 51 0 (4): 0 1063--1119, 2013

    work page 2013

  61. [61]

    D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning representations by back-propagating errors. Nature, 323 0 (6088): 0 533--536, 1986

    work page 1986

  62. [62]

    Salinas, V

    D. Salinas, V. Flunkert, J. Gasthaus, and T. Januschowski. DeepAR : Probabilistic forecasting with autoregressive recurrent networks. International Journal of Forecasting, 36 0 (3): 0 1181--1191, 2020

    work page 2020

  63. [63]

    A. A. Salisu, R. Gupta, and W. J. Kim. Exchange rate predictability with nine alternative models for BRICS countries. Journal of Macroeconomics, 71: 0 103374, 2022

    work page 2022

  64. [64]

    T. B. Sara c and K. Karag \"o z. Impact of short-term interest rate on exchange rate: The case of Turkey . Procedia Economics and Finance, 38: 0 195--202, 2016

    work page 2016

  65. [65]

    S. L. Scott and H. R. Varian. Predicting the present with Bayesian structural time series. International Journal of Mathematical Modelling and Numerical Optimisation, 5 0 (1-2): 0 4--23, 2014

    work page 2014

  66. [66]

    Sengupta, T

    S. Sengupta, T. Chakraborty, and S. K. Singh. Forecasting CPI inflation under economic policy and geopolitical uncertainties. International Journal of Forecasting, 41 0 (3): 0 953--981, 2025

    work page 2025

  67. [67]

    C. Y. Sin. The economic fundamental and economic policy uncertainty of Mainland China and their impacts on Taiwan and Hong Kong . International Review of Economics & Finance, 40: 0 298--311, 2015

    work page 2015

  68. [68]

    Stoica and I

    O. Stoica and I. Ihnatov. Exchange rate regimes and external financial stability. Economic Annals, 61 0 (209): 0 27--43, 2016

    work page 2016

  69. [69]

    J. B. Taylor. The role of the exchange rate in monetary-policy rules. American Economic Review, 91 0 (2): 0 263--267, 2001

    work page 2001

  70. [70]

    Trapletti, F

    A. Trapletti, F. Leisch, and K. Hornik. On the ergodicity and stationarity of the ARMA (1,1) recurrent neural network process. Technical report, SFB Adaptive Information Systems and Modelling in Economics and Management Science, WU Vienna University of Economics and Business, 1999

    work page 1999

  71. [71]

    Trapletti, F

    A. Trapletti, F. Leisch, and K. Hornik. Stationary and integrated autoregressive neural network processes. Neural Computation, 12 0 (10): 0 2427--2450, 2000

    work page 2000

  72. [72]

    V. Vovk, A. Gammerman, and G. Shafer. Algorithmic Learning in a Random World. Springer, 2005

    work page 2005

  73. [73]

    Wieland and M

    V. Wieland and M. Wolters. Forecasting and Policy Making . In G. Elliott and A. Timmermann, editors, Handbook of Economic Forecasting, volume 2 of Handbook of Economic Forecasting, pages 239--325. Elsevier, 2013

    work page 2013

  74. [74]

    L. Xu, S. Zhang, and R. Zhang. A note on Foster–Lyapunov drift condition for recurrence of markov chains on general state spaces. Journal of Theoretical Probability, 31 0 (4): 0 1923--1928, 2018

    work page 1923

  75. [75]

    A. Zeng, M. Chen, L. Zhang, and Q. Xu. Are transformers effective for time series forecasting? In Proceedings of the Thirty-Seventh AAAI Conference on Artificial Intelligence and Thirty-Fifth Conference on Innovative Applications of Artificial Intelligence and Thirteenth Symposium on Educational Advances in Artificial Intelligence. AAAI Press, 2023

    work page 2023

  76. [76]

    Z. Zhou, Z. Fu, Y. Jiang, X. Zeng, and L. Lin. Can economic policy uncertainty predict exchange rate volatility? New evidence from the GARCH-MIDAS model. Finance Research Letters, 34: 0 101258, 2020

    work page 2020