This site uses cookies.
Some of these cookies are essential to the operation of the site,
while others help to improve your experience by providing insights into how the site is being used.
For more information, please see the ProZ.com privacy policy.
VIA AILooking for an interpreter?
Remoto by ProZ.com connects you with an identity-verified interpreter by video or phone in about 60 seconds — 300+ languages, on demand or booked ahead.
Try Remoto »
Member since May '22
Working languages:
English to Spanish German to Spanish Spanish (monolingual) English (monolingual) German (monolingual)
Jul 19, 2025 (posted viaProZ.com): Un pequeño artículo sobre errores en la traducción de Stardew Valley: https://www.linkedin.com/pulse/la-mala-traducci%25C3%25B3n-al-espa%25C3%25B1ol-de-stardew-valley-salvador-balbuena-djmec...more, + 9 other entries »
Freelance translator and/or interpreter, Verified member
Data security
This person has a SecurePRO™ card. Because this person is not a ProZ.com Plus subscriber, to view his or her SecurePRO™ card you must be a ProZ.com Business member or Plus subscriber.
Affiliations
This person is not affiliated with any business or Blue Board record at ProZ.com.
Services
Translation, Interpreting, Editing/proofreading, Subtitling, Training, AI + Human review (MT post-editing), Transcription, Terminology management, Linguistic validation, AI training
Expertise
Specializes in:
Physics
Mathematics & Statistics
Astronomy & Space
Engineering (general)
Engineering: Industrial
Economics
Finance (general)
Science (general)
Investment / Securities
Safety
Mining & Minerals / Gems
Metallurgy / Casting
Materials (Plastics, Ceramics, etc.)
Aerospace / Aviation / Space
Petroleum Eng/Sci
Human Resources
Social Science, Sociology, Ethics, etc.
Marketing
Internet, e-Commerce
Medical: Cardiology
Computers: Software
Computers: Systems, Networks
Electronics / Elect Eng
Energy / Power Generation
IT (Information Technology)
Media / Multimedia
Nuclear Eng/Sci
Architecture
Linguistics
Botany
Accounting
Business/Commerce (general)
Insurance
Management
Advertising / Public Relations
Cinema, Film, TV, Drama
Printing & Publishing
Gaming/Video-games/E-sports
Automotive / Cars & Trucks
International Org/Dev/Coop
Blockchain/Cryptocurrencies
Education / Pedagogy
Names (personal, company)
Mechanics / Mech Engineering
Computers: Hardware
Environment & Ecology
Construction / Civil Engineering
Metrology
Computers (general)
Automation & Robotics
Waste management/Waste disposal/Recycling
Chemistry; Chem Sci/Eng
Water resources management
Law: Contract(s)
Real Estate
Transport / Transportation / Shipping
Safety Data Sheets/Regulations
Market Research
Forestry / Wood / Timber
Other
Ships, Sailing, Maritime
Meteorology
Gambling/Casino/Bets
Also works in:
Mining & Minerals / Gems
Metallurgy / Casting
Materials (Plastics, Ceramics, etc.)
Aerospace / Aviation / Space
Petroleum Eng/Sci
Human Resources
Social Science, Sociology, Ethics, etc.
Marketing
Internet, e-Commerce
Medical: Cardiology
Computers: Software
Computers: Systems, Networks
Electronics / Elect Eng
Energy / Power Generation
IT (Information Technology)
Media / Multimedia
Nuclear Eng/Sci
Architecture
Linguistics
Botany
Accounting
Business/Commerce (general)
Insurance
Management
Advertising / Public Relations
Cinema, Film, TV, Drama
Printing & Publishing
Gaming/Video-games/E-sports
Automotive / Cars & Trucks
International Org/Dev/Coop
More
Less
Volunteer / Pro-bono work
Open to considering volunteer work for registered non-profit organizations
Spanish to English: Distribución de eigenvalores en simulaciones de trayectorias: una aplicación de la ley de Marčenko–Pastur General field: Science Detailed field: Mathematics & Statistics
Source text - Spanish Distribución de eigenvalores en simulaciones de trayectorias: una aplicación de la ley de Marčenko–Pastur
Una vez obtenidos los eigenvalores, nuevamente se les grafica y se muestran los primeros valores en una tabla. En este caso, sin embargo, la evidencia de memoria debería ser suficiente para realizar pruebas que demuestren que esta persistencia en efecto coincide con la propuesta por \textcite{mehta_random_2004} al utilizar la ley de Marčenko-Pastur. Para el caso de matrices aleatorias, a las cuales, debido a la naturaleza de los datos, se sabe pertenece esta simulación de trayectorias, una matriz de covarianza se puede definir como
con $\mathbf{X}$ y $\mathbf{X}^*$ las trayectorias a investigar \parencite{bai}. Para evaluar esta ley, se asume un conjunto de datos aleatorios, con distribución normal, media 0 y desviación estándar $\epsilon$, es decir, $\mathcal{N}(0, \epsilon^2)$. Aquí, $T$ representa número de observaciones, mientras que $D$ el total de dimensiones, es decir, aspectos como la cantidad de variables involucradas. Así, la ley de Marčenko-Pastur establece que en $T \longrightarrow \infty$, cuando $y = \frac{T}{Q}$ es constante, en la distribución ya mencionada, se tendría la función de densidad de probabilidad
con $b = \left(1 + \sqrt{y}\right)^2$ y $a = \left(1 - \sqrt{y}\right)^2$
Lo que esto implica es que una distribución $\mathcal{N}(0, \epsilon^2)$ debería tener valores dentro de ciertos rangos. Si es el caso para un conjunto de datos, es decir, si esta distribución coincide con la distribución de eigenvalores de un conjunto de datos, entonces esta no posee persistencia, ni otros valores que podrían asumirse en el MBF.
Para este, una superposición con los eigenvalores debería mostrar outliers, los cuales aparecerían debido a efectos de memoria en las colas pesadas.
Translation - English Eigenvalue distribution in trajectory simulations: an application of the Marčenko–Pastur law
Once the eigenvalues are obtained, they are plotted again and the first values are shown in a table. In this case, however, the evidence of memory should be sufficient to justify statistical testing to show that this persistence does indeed coincide with that proposed by \textcite{mehta_random_2004} using the Marčenko-Pastur law. In the case of random matrices, which, given the nature of the data, this simulation of trajectories is known to belong to, a covariance matrix can be defined as
with $\mathbf{X}$ and $\mathbf{X}^*$ the trajectories to investigate \parencite{bai}. To evaluate this law, a random dataset with normal distribution, mean 0 and standard deviation $\epsilon$, i.e., $\mathcal{N}(0, \epsilon^2)$ is assumed. Here, $T$ represents the number of observations, while $D$ represents the total of dimensions, i.e. the number of variables involved. Thus, the Marčenko-Pastur law states that in $T \longrightarrow \infty$, when $y = \frac{T}{Q}$ is constant, in the already mentioned distribution, we would have the probability density function
with $b = \left(1 + \sqrt{y}\right)^2$ and $a = \left(1 - \sqrt{y}\right)^2$
What this implies is that a distribution $\mathcal{N}(0, \epsilon^2)$ should have values within certain ranges. If this is the case for a dataset, that is, if this the eigenvalue distribution matches the expected range of the distribution of eigenvalues in a dataset, then it has no persistence, nor other values that could be assumed in the MBF.
For this, an overlay with eigenvalues should show outliers, which would appear due to memory effects in heavy tails.
Spanish to English: Estacionalidad en series de tiempo: prueba de Dickey-Fuller aplicada con R General field: Science Detailed field: Mathematics & Statistics
Source text - Spanish Estacionalidad en series de tiempo: prueba de Dickey-Fuller aplicada con R
Uno de los supuestos más importantes en el caso de las series de tiempo es el de la estacionalidad. Es importante que una serie de tiempo sea estacional, puesto que de esta manera se elimina una parte importante del ruido, se identifican patrones periódicos y es más fácil detectar cambios o alteraciones relevantes. En este sentido, se trata de uno de los supuestos más importantes en el manejo de series de tiempo, por lo que cuando una serie de tiempo no es estacional (cosa que ocurre en prácticamente todos los casos), se realizan transformaciones que garanticen la estacionalidad.
Este supuesto puede romperse en el caso de las distribuciones no gaussianas, lo que implica que modelos típicos como los ARIMA o GARCH no pueden usarse, pero, lo que es más, que tampoco modelos que involucren persistencia tales como los ARFIMA o FIGARCH. Por ello, es necesario realizar pruebas, como la de Dicke-Fuller, a través de la búsqueda de raíces unitarias (que delatan a la estacionalidad), trata de establecer si en efecto la serie tiene esta propiedad \parencite{gujarati_basic_2009}.
Dickey-Fuller usa un modelo de regresión sencillo de orden 1 tal que $y_t = \rho y_{t-1} + \varepsilon_t$, con $\varepsilon_t$ el término estocástico en el momento $t$, $\rho$ el coeficiente de la regresión y $y$ la variable.
Para esta serie de tiempo la hipótesis nula es que no existe estacionalidad en la serie. Por tanto, un p-valor menor a 0.01 indicaría que hay razón estadística para rechazarla y optar por la hipótesis alternativa, que existe estacionalidad.
En el caso de la función \texttt{adf.test} de la biblioteca \texttt{tseries} \parencite{trapletti_tseries_1999} realiza estas regresiones en distintas ventanas de tiempo, obteniendo valores p para cada una de ellas.
En este caso, tal como la tabla \ref{tab:dickefuller} muestra, en todos los casos se muestra que es necesario aceptar la hipótesis alternativa, es decir, que la serie de retornos logarítmicos es ya estacionaria. Para la tabla \ref{tab:dickefuller}, se muestran solo los casos del primer y último valor.
Translation - English Seasonality in time series: Dickey-Fuller test applied with R
One of the most important assumptions in the case of time series is that of seasonality. It is important the presence of seasonality, this way a significant part of the noise is removed, periodic patterns are identified and it is easier to detect relevant changes or alterations. In this sense, it is one of the most important assumptions in the management of time series, so when a time series is not seasonal (which happens in practically all cases), transformations are applied that guarantee seasonality.
This assumption can be broken in the case of non-Gaussian distributions, implying that typical models such as ARIMA or GARCH cannot be used, and, what is more, neither models involving persistence such as ARFIMA or FIGARCH. Therefore, it is necessary to carry out tests, such as Dicke-Fuller, which, through the search for unit roots (which show seasonality), tries to establish if the series has this property \parent{gujarati_basic_2009}.
Dickey-Fuller uses a simple regression model of order 1 such that $y_t = \rho y_{t-1} + \varepsilon_t$, with $\varepsilon_t$ the stochastic term at the time $t$, $\rho$ the regression coefficient, and $y$ the variable.
For this time series, the the null hypothesis assumes the series is non-stationary in the series. Therefore, a p-value less than 0.01 would indicate that there is a statistical reason to reject it and opt for the alternative hypothesis, that there is seasonality.
In the case of the \texttt{adf.test} function of the \texttt{tseries} library \parencite{trapletti_tseries_1999} performs these regressions in different time windows, obtaining p-values for each of them.
In this case, as the table \ref{tab:dickefuller} shows, in all cases it is shown that it is necessary to accept the alternative hypothesis, i.e., that the series of logarithmic returns is already stationary. For the \ref{tab:dickefuller} table, only the cases of the first and last test cases are displayed.
English to Spanish: Muestra de traducción técnica (no oficial) - Mu-Checker (Mitutoyo) General field: Tech/Engineering Detailed field: Materials (Plastics, Ceramics, etc.)
Source text - English Note on stylus angle
If the stylus of a pivot bearing type probe (No. 519-521, No. 519-522, or
No. 519-327) makes an angle with a workpiece surface, as in the figure,
calibration should be performed for accurate measurement. Alternatively,
the displayed value may be corrected by multiplying it by the appropriate
correction factor as given in the table.
Common Specifications
Indication accuracy
±1%/± full scale
Analog output accuracy
±0.1%± within full scale (excluding the probe)
Analog output
±1V/± full scale
Meter indication response
Approx. 0.3s
Zero-setting enabled zone
1/3 scale or less (CMP mode)
Zero point thermal coefficient
100ppm/°C or less
Sensitivity thermal coefficient
100ppm/°C or less
Power consumption
5VA or less
Operating temperature range (ºC) 0 to 40
Storage temperature range (ºC)
-10 to 50
Power supply
120VAC
External dimensions (mm)
134 (W) ×210 (H) ×183 (D)
Measuring range (μm)
±1500, 500, 150, 50, 15, 5
Minimum reading (μm)
50, 10, 5, 1, 0.5, 0.1
Measuring range (in)
±.05, ±.015, ±.005, ±.0015, ±.0005, ±.00015
Minimum reading (in)
.001", .0005", .0001, .000050", .000010", .000005"
Features
• The EV-16A allows 6 probes to be connected unit simultaneously. one unit.
The use of the RS link enables easy construction of a multi-point measuring
system that allows a maximum of 10 units (60 probes).
• I/O outputs such as RS-232C, BCD, tolerance judgment and segment are selectable.
• Peak measurement and arithmetic operation between axes (in the same unit)
are also possible in addition to normal measurement for each axis.
Major Functions
• External control (zero-setting, presetting, etc.)
• Direction switching
• Error messaging
• Tolerance judgment output
• Various data output (RS-232C, BCD, segment)
• Peak measurement (maximum value, minimum value, runout) and arithmetic
operation (addition, average, maximum value, minimum value, maximum
width) between axes
Translation - Spanish Nota sobre el ángulo del palpador
Si el palpador de una sonda con cojinete de pivote (n.º 519-521, n.º 519-522, o
519-327) hace un ángulo con una superficie de la pieza de trabajo, como en la figura,
la calibración debe realizarse para una medición precisa. Alternativamente,
el valor mostrado puede corregirse multiplicándolo por factor
de corrección apropiado, tal como se indica en la tabla.
Especificaciones comunes
Precisión de la indicación
±1%/± escala completa
Precisión de salida analógica
±0,1%± dentro de la escala completa (excluyendo la sonda)
Salida analógica
±1V/± escala completa
Respuesta de indicación del medidor
Aprox. 0,3 s
Zona habilitada para el ajuste a cero
Escala 1/3 o menos (modo CMP)
Coeficiente térmico de punto cero
100 ppm/°C o menos
Coeficiente térmico de sensibilidad
100 ppm/°C o menos
Consumo de energía
5 VA o menos
Rango de temperatura de funcionamiento (ºC) 0 a 40
Rango de temperatura de almacenamiento (ºC)
De -10 a 50
Fuente de alimentación
120 VAC
Dimensiones externas (mm)
134 (An) ×210 (Al) ×183 (P)
Rango de medición (μm)
±1500, 500, 150, 50, 15, 5
Lectura mínima (μm)
50, 10, 5, 1, 0.5, 0.1
Rango de medición (pulgadas)
±.05, ±.015, ±.005, ±.0015, ±.0005, ±.00015
Lectura mínima (pulgadas)
.001", .0005", .0001, .000050", .000010", .000005"
Funciones
• El EV-16A permite conectar 6 sondas de forma simultanea.
El uso del enlace RS permite construir fácilmente un sistema de medición multipunto
que permite un máximo de 10 unidades (60 sondas).
• Se pueden seleccionar salidas de E/S COMO RS-232C, BCD, juicio de tolerancia y segmento.
• Medición de picos y operación aritmética entre ejes (en la misma unidad)
también son posibles además de la medición normal para cada eje.
Funciones principales
• Control externo (puesta a cero, preajuste, etc.)
• Cambio de dirección
• Mensajes de error
• Salida de juicio de tolerancia
• Salida de datos varios (RS-232C, BCD, segmento)
• Medición de picos (valor máximo, valor mínimo, excentricidad) y operaciones
aritméticas (suma, promedio, valor máximo, valor mínimo, máximo
ancho) entre ejes
More
Less
Experience
Years of experience: 12. Registered at ProZ.com: Apr 2022. Became a member: May 2022.
Get help on technical issues / improve my technical skills
Learn more about additional services I can provide my clients
Learn more about the business side of freelancing
Find a mentor
Stay up to date on what is happening in the language industry
Help or teach others with what I have learned over the years
Transition from freelancer to agency owner
Improve my productivity
Bio
I’m a technical translator with a strong background in engineering physics and applied statistics. My work focuses on translating complex, high-density content into clear, accurate communication for specialists in academia and industry. I’ve translated theses, papers, and documentation related to time series models (ARFIMA, FIGARCH), mining systems, automotive technologies, and stochastic modeling of real-world processes. Thanks to my academic training and hands-on experience in data-driven modeling, I can handle technical concepts with precision, bridging the gap between the source material and the target audience. I work with CAT tools like Trados and MemoQ to ensure consistency and speed, while always tailoring the output to the context and audience.
This user has earned KudoZ points by helping other translators with PRO-level terms. Click point total(s) to see term translations provided.
Member since May '22
Expertise 
