A novel forecasting model for the Baltic dry index utilizing optimal squeezing

• Author: Andreas Merikas,Spyros Makridakis,Mike G. Tsionas,Marwan Izzeldin,Anna Merika
• DOI: http://doi.org/10.1002/for.2613
• Published date: 01 January 2020
• Date: 01 January 2020

RESEARCH ARTICLE

A novel forecasting model for the Baltic dry index utilizing

optimal squeezing

Spyros Makridakis

1

| Andreas Merikas

2

| Anna Merika

3

| Mike G. Tsionas

4

|

Marwan Izzeldin

4

1

Institute For the Future (IFF), University

of Nicosia, Nicosia, Cyprus

2

Department of Maritime Studies,

University of Piraeus, Piraeus, Greece

3

Department of Economics, Deree

College, The American College of Greece,

Athens, Greece

4

Lancaster University Management

School, Lancaster, UK

Correspondence

Spyros Makridakis, Institute For the

Future (IFF), University of Nicosia, 46

Makedonitissa Av. Nicosia 2417, Cyprus.

Email: spyros.makridakis@gmail.com

Abstract

Marine transport has grown rapidly as the result of globalization and sustain-

able world growth rates. Shipping market risks and uncertainty have also

grown and need to be mitigated with the development of a more reliable pro-

cedure to predict changes in freight rates. In this paper, we propose a new fore-

casting model and apply it to the Baltic Dry Index (BDI). Such a model

compresses, in an optimal way, information from the past in order to predict

freight rates. To develop the forecasting model, we deploy a basic set of predic-

tors, add lags of the BDI and introduce additional variables, in applying Bayes-

ian compressed regression (BCR), with two important innovations. First, we

include transition functions in the predictive set to capture both smooth and

abrupt changes in the time path of BDI; second, we do not estimate the param-

eters of the transition functions, but rather embed them in the random search

procedure inherent in BCR. This allows all coefficients to evolve in a time‐

varying manner, while searching for the best predictors within the historical

set of data. The new procedures predict the BDI with considerable success.

KEYWORDS

Baltic dry index, Bayesian methods, compressed regression, forecasting, maritime shipping

1|INTRODUCTION

The Baltic Dry Index (BDI) has grown into a global eco-

nomic indicator, a mirror of world trade, whose com-

bined direct and indirect impact on the world economy

contributes, through the operation of merchant ships, to

about US $380 billion in freight rates, which is equivalent

to around 5% of global trade according to UNCTAD

(2015). As a weighted average of time charter freight

rates, BDI reflects both the supply of cargo ships and

the demand for transporting raw and other materials.

Until recently, its low levels indicated vessel overcapacity,

as well as a slowdown in demand for dry bulk commodi-

ties. Since mid‐2017 the index has shown definite signs of

recovery.

In 2008, the index reached 12,000 points, the highest

level in its history; since then, it has entered a spiraling

fall, registering a low of 290 in the first quarter of 2016

(Bloomberg, 2016). Yet, the purchase and sale activity of

vessels both in 2014 and 2015, according to Clarkson's

research (2016), exceeded the 2008 levels, with Europeans

being net purchasers and Asians net sellers. This could

indicate a turning point, at which economic activity

may be entering a new phase of the shipping cycle.

Shipping decisions on the part of ship owners and

charterers alike depend on the expected fluctuations of

the BDI. Decisions involve entering charter contracts of

different durations, switching between spot and time

charter operations, improving hedging performance using

derivative contracts, as well as whether to invest in newly

Received: 27 September 2018 Accepted: 10 May 2019

DOI: 10.1002/for.2613

Journal of Forecasting. 2020;39:56–68.wileyonlinelibrary.com/journal/for© 2019 John Wiley & Sons, Ltd.56