Your CFO has a few questions for us:
How do we characterize semiconductor variability and the impact of one market on another?
What are the best combinations of semiconductor drivers?
How much capital is needed to support a semiconductor earnings stream?
How should the company plan to meet risk tolerances and thresholds for losses?
For the semiconductor sector we select exchange traded funds (ETF) from the semiconductor sector:
USD
PSI
SMH
These funds act as indices to effectively summarize the inputs, process, management, decisions, and outputs of various aspects of the semiconductor sector.
We load historical data on three ETFs, tranform prices into returns, and then further transform the returns into within-month correlations and standard deviations.
Our process includes
Review the stylized facts of volatility and relationships among three repesentative markets.
Develop market risk measures for each driver of earnings.
Apply corporate risk tolerances and thresholds to determine optimal collateral positions for each driver of earnings.
Determine optimal combinations of the drivers for maximum excess portfolio return relative to portfolio risk as well as the minimization of risk
Given the optimal maximum excess return per risk portfolio, determine the probable range of collateral needed to satisfy corporate risk tolerance and thresholds.
Within the Semiconductor market segment, manufacturers are developing technologies to support a wide variety of end use applications such as data processing, wireless, consumer, and automotive products. Exchange-traded funds (ETF) are heavily dependent on the end use market segments which are supported by the semiconductor suppliers within each ETF. The end use applications which are generating the most revenue in these segments are data center servers, solid state drives, mobile pc, and smartphone technologies.
Stylized Facts:
Moore’s Law states that the number of transistors in a dense integrated circuit doubles about every two years.
Semiconductor revenue growth is strongly correlated to global economic growth.
Over the last several years there has been a significant need for large amounts of data storage and processing capability. Data generation is occurring at an astounding rate. The exponential growth is requiring cloud storage suppliers to continually acquire new equipment and/or provider supplied services. The trend is will continue to grow exponentially in order to support the increase in data storage and processing that is required to meet these needs.
Stylized Fact:
Data generation will continue to grow exponentially as an abundance of data generating platforms are offered to consumers. These platforms include social media, communication, digital photos, services, and “internet of things” connected devices.
Wireless devices are now being developed with advanced features such as biometrics, artificial intelligence, and increased battery life. Some examples of these features include fingerprint readers, facial recognition, light sensors, MEMS microphones, and motion sensors. Integration of these features requires a combination of high battery capacity, engineered semiconductors and integrated circuits with low power consumption. These technologies require the development of novel integrated circuits, and semiconductor materials that enable tech companies to incorporate these features into their latest and greatest high-end devices. In 2018, consumers did not purchase high-end smart phone models as abundantly as seen in prior years. The mid-tier smartphone technologies yielded must higher growth than top-of-the-line models since of these features are now several years old, and the cost of incorporation has declined significantly. Even though there was a slight decline in revenue generated from high-end consumer mobile devices, there was still growth in this segment. This decline is most likely attributed to a lack of newly developed features from the prior year’s mobile device models.
Stylized Fact:
Smart phone developers will continue to push the limits of their devices. There will always be an interest in improving upon existing features, especially battery life. The need for chips with a low power consumption footprint will continue to drive development of new semiconductor products and manufacturing methods.
The consumer market is the primary driver within the semiconductor industry. Growth is primarily due to advancements in internet connected devices. In recent years, there has been an abundance of new devices for smart home applications. These new devices require the ability to connect to a wireless gateway, thus requiring additional ICs that enable this feature. Consumer adoption of these devices is increasing as their cost to the consumer decreases. The need for DRAM and NAND memory within these devices is also yielding strong growth in ETFs that have a strong presence in memory manufacturing.
Stylized Fact:
The need for more advanced integrated circuits is necessary for to incorporate wireless capability of “internet of things” devices such as smart home devices, wearables, and otherwise non-connected devices.
The need for advanced semiconductors in automotive applications is becoming more prevalent as automakers are developing advanced safety features in vehicles such as adaptive cruise control (ACC), emergency braking (AEB), lane departure warning, blind spot detection, rear-view camera, safety exit assist, and facial recognition. Just like the other segments mentioned above, these features require more advanced sensors that can only be enabled by the semiconductor technology developed in the last ten years. At first these technologies were only available in high-end automobiles, but further adoption has driven costs down for semiconductor suppliers due to economies of scale.
Stylized Fact:
Smart connectivity and advanced safety features will be adopted by all automobile manufacturers as the cost continues to decline due to economies of scale.
PSI SMH USD
PSI 1.0000000 0.9829207 0.9860088
SMH 0.9829207 1.0000000 0.9930022
USD 0.9860088 0.9930022 1.0000000
| mean | median | std_dev | IQR | skewness | kurtosis | |
|---|---|---|---|---|---|---|
| SMH.size | 1.3260 | 0.9928 | 1.2242 | 1.2455 | 1.7508 | 6.4578 |
| USD.size | 2.6566 | 1.9122 | 2.4560 | 2.5967 | 1.7351 | 6.3694 |
| PSI.dir | -0.0803 | -1.0000 | 0.9968 | 2.0000 | 0.1612 | 1.0301 |
| SMH.dir | -0.0361 | -1.0000 | 1.0014 | 2.0000 | 0.0723 | 1.0052 |
[1] 1.350853
References: 5 Top Semiconductor ETFs, 2017, The Motley Fool, Found at: https://www.fool.com/investing/2017/07/21/5-top-semiconductor-etfs.aspx
Biometric Sensor Chips Information, 2019, IEEE GlobalSpec, Found at: https://www.globalspec.com/learnmore/semiconductors/sensor_ic/biometric_sensors
Global Semiconductor Market Trends, 2018, HIS Markit, Found at: http://theconfab.com/wp-content/uploads/P-18-Len-Jelinek.pdf
Semiconductor Industry, 2019, Wikipedia, Found at: https://en.wikipedia.org/wiki/Semiconductor_industry
The Industry Handbook: The Semiconductor Industry, 2018, Investopedia, Found at: https://www.investopedia.com/features/industryhandbook/semiconductor.asp
The Top 5 Semiconductor ETFs, 2018, Investopedia, Found at: https://www.investopedia.com/articles/etfs/top-semiconductor-etfs/
The Top 7 Car Safety Features You Canot Do Without, 2018, Forbes, Found at: https://www.forbes.com/sites/peterlyon/2018/05/31/the-top-7-car-safety-features-you-cannot-do-without/#2f2e87115fc0