{"paper_id":"2402.07435","title":"Analyzing Currency Fluctuations: A Comparative Study of GARCH, EWMA, and IV Models for GBP/USD and EUR/GBP Pairs","authors":["Narayan Tondapu"],"paper_date":"2024-02-12","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2402.07435v1","page_url":"https://thequant.space/flowcharts/analyzing-currency-fluctuations-a-comparative-study-of-garc/","scores":{"empirical_rigor":8,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a comparative analysis of EWMA, GARCH‑type, and implied volatility models to forecast 20‑day volatility for GBP/USD and EUR/GBP, identifying the best‑performing specifications.","quote":"Our experiments reveal that for the GBP/USD pair, the most accurate volatility forecasts stem from the utilization of GARCH models employing a rolling window methodology.","summary":"In this study, we examine the fluctuation in the value of the Great Britain Pound (GBP). We focus particularly on its relationship with the United States Dollar (USD) and the Euro (EUR) currency pairs. Utilizing data from June 15, 2018, to June 15, 2023, we apply various mathematical models to…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Residuals follow a t‑distribution rather than a normal distribution.","type":"statistical","stated":"implied","quote":"However, since returns data may better fit a t‑distribution, the Quasi‑MLE algorithm is often preferred, relaxing the normality assumption as shown in Equation (3)."},{"assumption":"Implied volatility data are scaled assuming 252 trading days per year.","type":"data","stated":"explicit","quote":"Typically, this data is scaled down by a factor of 252, assuming 252 trading days in a year, where n represents the number of days for forecasting volatility."},{"assumption":"EWMA model uses a decay factor of 0.97.","type":"model","stated":"explicit","quote":"In this investigation, we've scrutinized and evaluated various methods"}]},"datasets":{"status":"available","items":[{"name":"GBP/USD and EUR/GBP daily returns","provider":null,"kind":"market-prices","access":"unknown","period":"June 15, 2018 to June 15, 2023","universe":"GBP/USD and EUR/GBP currency pairs","quote":"Utilizing data from June 15, 2018, to June 15, 2023, we apply various mathematical models to assess their effectiveness in predicting the 20‑day variation in the pairs' daily returns."},{"name":"Annualized implied volatility","provider":"Bloomberg Terminal","kind":"options","access":"commercial","period":null,"universe":"GBP/USD and EUR/GBP pairs","quote":"Initially, annualized implied volatility data for each asset is obtained from the Bloomberg Terminal.","known_id":"bloomberg"}],"known":[{"id":"bloomberg","name":"Bloomberg","provider":"Bloomberg L.P.","kind":"market-prices","access":"commercial","quote":"Following this approach, we formulate regression equations to depict volatility based on the annualized data sourced from Bloomberg as shown in Equation (7). 𝜎𝑡 2 = 𝛽𝑜+ 𝛽1𝜎 𝑖𝑚𝑝𝑙𝑖𝑒𝑑, 𝑡−1 + 𝛽2𝜎 𝑎𝑐𝑡𝑢𝑎𝑙, 𝑡−1 + 𝜇𝑡 (7) The regression model is trained using the initial 898 data points, constituting the training subset.","mentions":2,"used_in":2},{"id":"yahoo-finance","name":"Yahoo Finance","provider":"Yahoo","kind":"market-prices","access":"public","quote":"Empirical Study 3.1 Data Analysis For our study we gathered data from Yahoo Finance, which included the daily market prices of currency pairs.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"Significant γ values in EGARCH, GJR‑GARCH and TGARCH models indicate asymmetric returns for EUR/GBP but not for GBP/USD.","kind":"statistical","quote":"significant γ values at various confidence intervals for EGARCH, GJR‑GARCH, and GJR‑GARCH (2,1,2) models indicate asymmetric returns. This contradicts prior findings..."},{"finding":"GJR‑GARCH models with rolling window forecasts provide the most accurate predictions for GBP/USD (RMSE 0.128, MAE 0.100).","kind":"predictive-accuracy","quote":"the most accurate forecast for GBP/USD originates from the GJR‑GARCH (2, 1 ,2) model, with an RMSE of 0.128, while for MAE, the GJR‑GARCH(1, 1, 1) model performs best with a value of 0.100."},{"finding":"For EUR/GBP, OLS regression models based on implied volatility outperform GJR‑GARCH in MAE (MAE 0.080).","kind":"predictive-accuracy","quote":"OLS regression models outperformed in terms of MAE."},{"finding":"Modeling GARCH residuals with a t‑distribution yields lower AIC and BIC scores than assuming normality.","kind":"statistical","quote":"modeling residual values in GARCH models with a t‑distribution yielded improved fit compared to a normal distribution, as evidenced by lower AIC and BIC scores across all tested models."},{"finding":"Rolling window and expanding window forecasts show no clear superiority across tested scenarios.","kind":"performance","quote":"forecasting techniques between Rolling Window methods did not yield a clear preference, contrary to initial expectations."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":26,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T18:14:56Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}