Macroeconomic forecasting for Pakistan in a data-rich environment

Syed, Ateeb Akhter Shah; Lee, Kevin Haeseung

doi:10.1080/00036846.2020.1826399

Cited by 6 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Instead, it published the YOY and MOM inflation series based on the CPI indices, available only from 2016:07 in the ED. Second, we use the QIM as a measure of output/GDP or output/GDP growth as almost all studies that have used monthly datasets to forecast inflations, or any other macroeconomic variable in a multivariate setting for Pakistan, have used the same variable to proxy for output/GDP [see Syed & Lee (2021), Hussain, et al (2018), Hussain & Mahmood (2017) and Hanif & Malik (2015)]. However, the base of this measure was revised from 2005-06 to 2015-16 by the PBS on January 19, 2022.…”

Section: Methodsmentioning

confidence: 99%

“…Studies have indicated that using machine-learning to forecast key macroeconomic variables such as inflation, output or GDP growth, and key interest rates, has become a recent and recurring trend [see Eickmeier & Ng (2011), Li & Chen (2014), Panagiotelis, et al (2019), Syed & Lee (2021), Mederios, et al (2021), among others]. These techniques are generally capable of handling a large amount of data and outperform the commonly used univariate and multivariate econometric models (inclusive of the dynamic factor model, which can handle a large amount of data) (Mederios, et al (2021).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

EasyData: A Monthly Dataset for Macroeconomic Research on Pakistan

2023

LJE

View full text Add to dashboard Cite

This paper introduces the monthly State Bank of Pakistans EasyData , for conducting empirical macroeconomic analysis and forecasting for Pakistan's economy. For this purpose W e perform a forecasting exercise using the conventional econometric models and the most recent machine learning algorithms. We find that the machine learning models outperform the benchmark and regression models based on observed factors. Furthermore, the dataset has a higher ability to predict the external variables, a possible outcome of Pakistan's economy and its persistent balance of payment problem. The focus of policy has been to address this issue.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

EasyData: A Monthly Dataset for Macroeconomic Research on Pakistan

2023

LJE

View full text Add to dashboard Cite

show abstract

“…In this paper, we use several classical econometric models taken from Hanif & Malik (2015) and compare their forecasting accuracy against the DFM and ML models. We follow Syed & Lee (2021) in explaining the general methodology for forecasting used in this paper. We initiate our analysis by using the simple mean model (naïve), the AR, the ARDL models, the structural VAR models followed by the DFM, and the sophisticated machine learning techniques such as the Ridge (Ridge), the LASSO, the elastic net (EN), artificial neural network (ANN) and the RF.…”

Section: Forecasting Methodsmentioning

confidence: 99%

Inflation Forecasting for Pakistan in a Data-rich Environment

Shah,

Muhammad Ishtiaq,

Sumbal Qureshi

et al. 2023

pdr

View full text Add to dashboard Cite

This paper uses machine learning methods to forecast the year-on-year CPI inflation of Pakistan and compare their forecasting performance against the comprehensive traditional forecasting suite contained in Hanif and Malik (2015). It also augments the comprehensive forecasting suite with the dynamic factor model which is able to handle a large amount of information and put all of these models in competition against the latest machine learning models. A set of 117 predictors covering a period of July 1995 to June 2020 is used for this purpose. We set the naïve mean model as the benchmark and compare its forecasting performance against 14 traditional and 5 sophisticated machine learning models. We forecast the year-on-year CPI inflation over a 24 months horizon. Forecasting performance is measured using the RMSE. Our results show that the machine learning approaches perform better than the traditional econometric models at 18 forecast horizons.

show abstract

“…It usually produces more accurate solutions than a single model would. They two most well-known methods in ensemble learning are bagging [16] This model is generated using the same machine learning algorithm with n random observations of m sub-samples of the original dataset using the bootstrap sampling method. The second step is aggregating the result generated from these models.…”

Section: A Ensemble Classifiermentioning

confidence: 99%

“…A false or inaccurate hypothesis can be improved if the training dataset for the learning algorithm is insufficient. The most well-known methods in ensemble learning are bagging and clustering [16]- [19] and boosting [20]. These algorithms achieve sufficient classification results and are commonly used to generate numerous ensemble models.…”

Section: Introductionmentioning

confidence: 99%

Proposed Hybrid Ensemble Learning Algorithms for an Efficient Intrusion Detection System

2022

IJCCCE

View full text Add to dashboard Cite

Due to sophisticated cyber-attacks, and to produce false alarms on suspicious or unusual behavior to monitor computer resources, Intrusion Detection Systems (IDSs) are required. Hence, Many Machine Learning (ML) and data mining techniques have been proposed to increase the effectiveness of IDSs, whereas current IDS algorithms are still struggling to perform effectively while many IDSs depend on a single classifier to detect intrusions. Single-classifier IDSs cannot achieve high accuracy and low false alarm rates because of zero-day attacks. In this paper, a hybrid ensemble method using AdaBoosting and Bagging for IDS is proposed. This study aims to identify unknown (zero-day attacks) and known (well-known) attacks. So, the proposed model comprises three stages. The first stage is preprocessing. The second stage involves the application of AdaBoosting and Bagging methods by four different classifiers modifying (i.e., Naïve Bayesian (NB), Support Vector Machine (SVM), random forest (RF), and K_Nearest Neighbor (KNN)). Such a modification is performed for the AdaBoosting methods. The AdaBoosting classifier is then combined to work in the Bagging method. For attack recognition, uses the voting technique as the third stage. Experimental results reveal that using the UNSW BN15 dataset yields testing with 85.49% accuracy, 99.96% detection rate, and 0.006 false alarm rate. Therefore, the proposed Hybrid AdaBoosting and Bagging Method (HABBM) can outperform other comparable and state-of-the-art techniques across a variety of parameters. Index Terms—AdaBoosting method, Bagging method, Cyber Security CS, Ensemble Method, Intrusion Detection Systems IDSs.

show abstract

Macroeconomic forecasting for Pakistan in a data-rich environment

Cited by 6 publications

References 17 publications

EasyData: A Monthly Dataset for Macroeconomic Research on Pakistan

EasyData: A Monthly Dataset for Macroeconomic Research on Pakistan

Inflation Forecasting for Pakistan in a Data-rich Environment

Proposed Hybrid Ensemble Learning Algorithms for an Efficient Intrusion Detection System

Contact Info

Product

Resources

About