Dr.P.Baba Gnanakumar

Gold & Gold Markets 2019

Predictive Analytics for Devising Gold Prices

A case study of Price prediction for Importing Gold from Thailand to India

Agenda

03

01Motivation of the Research

02 Aim of the Research

04Discussion

Analysis03

Motivation of the Analytics

2018:

The ministries

are looking at

the industry’s

demand of cutting

import duty on

gold to 4 per cent

from the current

10 per cent

Motivation of the Research

USP of Gold from Thailand

Customer Expectation

Customer Expectation (23 Carat)

Final price of the jewellery =

Price of (22 KT or 18 KT) gold X (W

eight in grams) +

Making charges +

GST at 3% on (Price of jewellery +

making charges)

Phenomenon in Pricing the Imported Gold

Traditional

method

Can we integrate the Retail

brand values of individual firms

with National / Global values

while pricing the Gold ?

What are the variables that shows the

“brand” values of Gold in

Regional / National / Global level ?

How we can price 23 Carat Gold

imported from other countries?

Core Area of Discussion

02 Aim of the Research

• To create distributed networks in estimating the Gold price and

thus enable to create a new platform for trading gold between India

and Thailand

• To fix the price of the Gold based on the demand trends and

• To set the automotive price fixing mechanism.

-404444444 Relationship exists in trend of Exchange rates and gold prices;

but no price discovery among the variables.

Bukowski (2016)

Dependency between the Gold prices among the Thailand,

Indonesia, and the Philippine (TIP) stock markets

Pastpipatkul (2016)

There exists relationship among gold prices in five global gold

markets, namely London, New York, Japan, Hong Kong and

Taiwan.

Chang (2013)

There exits relationship between gold prices and selected

financial indicators .

Omag (2012

Examined the economic theory of gold price movements in

India.

Abke (1980)

01

02

03

04

05

Knowledge Gap in

Literature Identification of Factors

influencing the Price of Gold

Literature Review

MethodologyStages in Research

Stage II

Stage III

Stage IV

Stage V

Stage VI

Stage I

Factors that are

influencing the gold

supply chain from the

past empirical research

findings

Empirically test the

relationship between

variables and gold price

with econometric modeling.

Tools are established

to predict the future

price of Gold

Price prediction of

Import of Gold to

Bangalore

Integrate the data warehouse

providers who are supplying

the value-based data for gold

supply chain based on the

above steps.

Modelling the

System

Data - Inputs to

Research

Indices from 1992 to

2017.

Bombay Stock

Exchange

Gold price from 1992 to

2017

World Gold Council

Exchange rate of Bhatt

from RBI

Exchange Rates

03

Past Research Work Variables affecting the

Price of GoldKannan, R., & Dhal, S. (2008). India's demand for gold: some issues for economic

development and macroeconomic policy. Indian Journal of Economics and Busines

s, 7(1), 107.

Interest rate, Exchange rate,

Vaidyanathan, A. (1999). Consumption of gold in India: Trends and determinants

. Economic and Political Weekly, 471-476.

Kaufmann, T. D., & Winters, R. A. (1989). The price of gold: A simple model. Resourc

es Policy, 15(4), 309-313.

Sjaastad, L. A. (2008). The price of gold and the exchange rates: Once again

. Resources Policy, 33(2), 118-124.

Exchange Rate

Cai, J., Cheung, Y. L., & Wong, M. (2001). What moves the gold market?

. Journal of Futures Markets, 21(3), 257-278.

Consumer products ‘ Index

Diba, B., & Grossman, H. (1984). Rational bubbles in the price of gold. Interest rate

Stage I

Stage I

Past Research Work Variables affecting the

Price of GoldShafiee, S., & Topal, E. (2010). An overview of global gold market and gold price for

ecasting. Resources Policy, 35(3), 178-189.

Inflation

Parisi, A., Parisi, F., & Díaz, D. (2008). Forecasting gold price changes: Rolling and

recursive neural network models. Journal of Multinational financial management

, 18(5), 477-487.

Historical Prices, (WGC)

Diba, B., & Grossman, H. (1984). Rational bubbles in the price of gold. Interest rate

Bhunia, A., & Mukhuti, S. (2013). The impact of domestic gold price on stock price

indices-An empirical study of Indian stock exchanges. Universal Journal of

Marketing and Business Research, 2(2), 35-43.

Patel, S. A. (2013). Causal relationship between stock market indices and gold

price: Evidence from India. IUP Journal of Applied Finance, 19(1), 99.

Capital Market’s Index

Stage I

Indian Gold Prices - Co-integration between Gold Price

in India Vs Prices in Thailand – Source : WGC (Price per troy ounce

of gold)

Country Specific Gold Price (Per Troy Ounce)

in Respective Currencies

Trace -Test Max- Eigen Test Significance based on tests

(out of two tests)

Indian Rupee & Thai baht 21.26 21.24 There is a co-integration

oy ounce

Correlation and Cointegration of Gold Price in India with BSE Indices

Index Correlation with

Gold Price

Co-integration with

Gold prices (Lag 2)

BSE - Sensex 0.917 No

BSE Consumer Durables Index 0.955 No

BSE Metal Index 0.729 No

BSE Capital Goods Index 0.929 No

BSE Consumer Discretionary Goods and Services Index 0.927 Yes

BSE FMCG Index 0.933 No

BSE Oil and Gas Index 0.923 No

BSE Realty Index -0.365 No

Variables influencing the Gold PricesData Base Requirement

Regional

Data base

World Gold

Council

Banks’

Data base BSE

Exchange Rate of

countries * (Having

predictable relationship)

----------------

BSE Consumer

Discretionary Goods.

-------------------

Gold Price in National

level.

--------------

Regional Price of the

Gold.

----------------------

Price Prediction

1. Import the libraries

and read the Gold

data

2. Define explanatory

variables

3. Define dependent

variable

4. Split the data to test

dataset

5. Create a regression

model

6. Predict the Gold

prices

Steps

Supervised

learning GARCH Modeling –

..Using R

Rate of change (ROC)

Regression Trend

Case Study :

Pricing for Gold Supply from Thailand

Transaction – Import of Gold from Thailand to Bangalore – Price Estimation

Gold Price in

India

5 yrs

World Gold Council

/ MCX (India)

Gold price derives

from National level

Gold Price in

Thailand

5 yrs

World Gold Council

The seller is from

Thailand

Exchange price

of Thai Bhat

5 yrs

RBI – Exchange

value between

India and Tahi bhatt

Exchange rate is

one of the

influencing factor

BSE Consumer

Discretionary

Goods

5 yrs

BSE

Values of stock

market affects the

Gold price

Gold Price in

Bangalore

5 yrs

Jewellers’ Association

of Bangalore

http://www.jab.org.in

The buyer of Gold

is from Bangalore

Exogenous

VariablesExogenous

Variables

Exogenous

Variables

Exogenous

Variables

Dependent

Variable

37

21.2 24.4

14.2

0

5

10

15

20

25

30

35

40

A1 A2 A3 A4

Gold Price in Bangalore

Gold Price in India

Exchange price of Thai Bhat

Gold Price in Thailand

BSE Consumer Discretionary Goods

Gold Price in Bangalore

Gold Price in India

Gold Price in Thailand

BSE Consumer Discretionary Goods

Gold Price in Indian Market (National)

Consumer Discretionary Goods Index

Weighted Exchange Rate of Thailand

Bhat on Gold Price

21.22

i in n BaBangngalalorore

22

1

Mean Absolute Percentage Error

Data Input – Validation of Different Alternatives

Simulation for determining number of Years

( Mean Absolute Percentage Error)

Output of -Case study

GARCH Equation

Co-efficient

0.931647.

Consumer Discretionary

Goods Index

Co-efficient

0.045742.

Gold Price in Indian

Market (National)

Co-efficient

-0.039368

Weighted Exchange Rate

of Thailand Bhat on Gold

Price

.

Coefficient Std. Error z-Statistic Prob.

C 3088.075 290.4721 10.63123 0.0000

RESID(-1)^2 0.623147 0.097879 6.366490 0.0000

GARCH(-1) -0.149531 0.051380 -2.910273 0.0036

MAE

=

66.11

MAPE

=

2.3%

R2

=

0.69

04Model

Estimation of Gold Price

95%

The results of the case study

predict the price to be quoted by

the exporter from Thailand with the

accuracy of 95 percent in the

particular transaction. .

Results

-600

-400

-200

0

200

400

2,200

2,400

2,600

2,800

3,000

3,200

3,400

II III IV I II III IV I II III IV I II III IV I II III IV I

2012 2013 2014 2015 2016

Residual Actual Fitted

The results of the case

prprprprededededicicict t t thththe e e prprpricicice e e tototo b b be e e ququququot

the exportrtrtererer f f f frorororom m m m ThThThaiaiailalalandndnd w

accuracycy of 95 p percent

95%

Model

Infographic StyleInsert the title of your subtitle Here

Get a modern

PowerPoint

Presentation that is

beautifully designed.

Your Text Here

Get a modern

PowerPoint

Presentation that is

beautifully designed.

Your Text Here

Model

Distributed

Databases

Transaction is committed to

Block chain and miners are

rewarded

Transaction is distrib

uted and validated

via cryptographic

hashing

Buyer creates a

transaction or a

block

Seller receives the transaction

Buyer

SellerCrypto-HashingBlock (Transaction)

+ Customs

Duty + Cost

of Carrying

Sl.No Variables Impact on Total MRP

(Variations on MRP)

1 Gold Price in Thailand 60.5 %

2 Demand Trends – Based on Currency variations

(Inverse Relation)

(+ or - ) 4.5%

3 Demand Trends – based on Capital Market Index (+ or -) 2%

4 Demand Trends – Regional / Festival variations (+ or -) 4%

5 Duty (GST +Import Duty, CVD, Customs etc)

(Tariff adjusted fortnightly)

+ 23 ( Max. 31%)

6 Brand Values + 3 %

7 Cost of Carrying & others + 2.5 %

Price Elasticity Components

Situations Price (on 26/1/2019) (INR)

Price of 24 Carat Gold of Thailand in Indian Rupee

(Open market)

29156 Per 10

grams

Importing Gold from Thailand (Whole Sale) 33928 ( Max.34346)

Buying through NRI & bringing less than 1kg Gold from

Bangkok

32071

Retail Price in India (24 carat in Bangalore) 32428

Customer based Price - Variations

Phenomenon in Pricing the Imported Gold

Policy Changes – Customer Experience

Flexible Pricing can be

possible with

Real Time Pricing

Mechanism

Elasticity of Demand.

Changes in Prices

Cha

nge

s in D

em

an

d

Based on Price and Demand

changes during August, 2018

Challenges in Research Overcoming the Challenges

Cascading Effect

Large Volume data

Cross data

Time constraint in estimation

Mismatch of Variables

Streaming and Mapping of Data

Challenges in Pricing

Automation

Integration

Automated machine algorithms must link the

regionally distributed ledger for identifying price at

the regional level. The semantic web ontology that

integrates the buyer and seller will address the

predictive pricing using Supervised learning

algorithm.

The way of integrating the variables with data

warehousing will create a transparent system of

price determination of gold in India.

Wrap Up

PricingThe Gold pricing will have multiple pricing based

on transaction values. The brand values are

considered in each stage of pricing.

Thank YouCreate, integrate & Apply

Let us transform the Gold pricing system

transparently

Analytics has the power to transform the Pricing system

CrCr